Publications of Andras Szilagyi

D-Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from a hyperthermophilic eubacterium, Thermotoga maritima, is remarkably heat stable (Tm = 109 degrees C). In this work, we have applied homology modelling to predict the 3-D structure of Th.maritima GAPDH to reveal the structural basis of thermostability. Three known GAPDH structures were used as reference proteins. First, the rough model of one subunit was constructed using the identified structurally conserved and variable regions of the reference proteins. The holoenzyme was assembled from four subunits and the NAD molecules. The structure was refined by energy minimization and molecular dynamics simulated annealing. No errors were detected in the refined model using the 3-D profile method. The model was compared with the structure of Bacillus stearothermophilus GAPDH to identify structural details underlying the increased thermostability. In all, 12 extra ion pairs per subunit were found at the protein surface. This seems to be the most important factor responsible for thermostability. Differences in the non-specific interactions, including hydration effects, were also found. Minor changes were detected in the secondary structure. The model predicts that a slight increase in alpha-helical propensities and helix-dipole interactions also contribute to increased stability, but to a lesser degree.

To reveal the structural basis of the increased thermal stability of 3-isopropylmalate dehydrogenase (IPMDH) from Thermus thermophilus, an extreme thermophile, the homology-based structural model of one mesophilic (Escherichia coli) counterpart, was constructed. Both IPMDHs are homodimeric proteins. We built a model of one subunit using the 3-D structures of the Th. thermophilus IPMDH and the homologous E.coli isocitrate dehydrogenase. Energy minimization and molecular dynamics simulated annealing were performed on the dimer, including a surrounding solvation shell. No serious errors were detected in the refined model using the 3-D profile method. The resulting structure was scrutinized and compared with the structure of the Th.thermophilus IPMDH. Significant differences were found in the non-specific interactions including the hydrophobic effect. The model predicts a higher number of ion pairs in the Th.thermophilus than in the E.coli enzyme. An increase was observed in the stabilities of alpha-helical regions in the thermophilic protein. The preliminary X-ray coordinates of the E.coli IPMDH were received after the completion of this work, allowing an assessment of the model in terms of the X-ray structure. The comparison proved that most of the structural features underlying the stability differences between the two enzymes were predicted correctly.

The leuB gene from the psychrotrophic strain Vibrio sp. I5 has been cloned and sequenced. The gene codes for 3-isopropylmalate dehydrogenase, a 360-residue, dimeric enzyme involved in the biosynthesis of leucine. Three recently solved homologous isopropylmalate dehydrogenase (IPMDH) crystal structures from thermophilic and mesophilic organisms have been used to build a homology model for the psychrotrophic IPMDH and to deduce the possible structural reasons for its decreased thermostability. According to our model the psychrotrophic IPMDH contains fewer stabilizing interactions than its mesophilic and thermophilic counterparts. Elements that have been identified as destabilizing in the comparison of the psychrotrophic, mesophilic and thermophilic IPMDHs are a smaller number of salt-bridges, a reduction in aromatic-aromatic interactions, fewer proline residues and longer surface loops. In addition, there are a number of substitutions of otherwise strictly conserved residues that can be linked to thermostability.

The comparison of the three-dimensional structures of thermophilic (Thermus thermophilus) and mesophilic (Escherichia coli) 3-isopropylmalate dehydrogenases (IPMDH, EC 1.1.1.85) suggested that the existence of extra ion pairs in the thermophilic enzyme found in the intersubunit region may be an important factor for thermostability. As a test of our assumption, glutamine 200 in the E. coli enzyme was turned into glutamate (Q200E mutant) to mimic the thermophilic enzyme at this site by creating an intersubunit ion pair which can join existing ion clusters. At the same site in the thermophilic enzyme we changed glutamate 190 into glutamine (E190Q), hereby removing the corresponding ion pair. These single amino acid replacements resulted in increased thermostability of the mesophilic and decreased thermostability of the thermophilic enzyme, as measured by spectropolarimetry and differential scanning microcalorimetry.

BACKGROUND: Proteins from thermophilic organisms usually show high intrinsic thermal stability but have structures that are very similar to their mesophilic homologues. From prevous studies it is difficult to draw general conclusions about the structural features underlying the increased thermal stability of thermophilic proteins. RESULTS: In order to reveal the general evolutionary strategy for changing the heat stability of proteins, a non-redundant data set was compiled comprising all high-quality structures of thermophilic proteins and their mesophilic homologues from the Protein Data Bank. The selection (quality) criteria were met by 64 mesophilic and 29 thermophilic protein subunits, representing 25 protein families. From the atomic coordinates, 13 structural parameters were calculated, compared and evaluated using statistical methods. This study is distinguished from earlier ones by the strict quality control of the structures used and the size of the data set. CONCLUSIONS: Different protein families adapt to higher temperatures by different sets of structural devices. Regarding the structural parameters, the only generally observed rule is an increase in the number of ion pairs with increasing growth temperature. Other parameters show just a trend, whereas the number of hydrogen bonds and the polarity of buried surfaces exhibit no clear-cut tendency to change with growth temperature. Proteins from extreme thermophiles are stabilized in different ways to moderately thermophilic ones. The preferences of these two groups are different with regards to the number of ion pairs, the number of cavities, the polarity of exposed surface and the secondary structural composition.

Some structural features underlying the increased thermostability of enzymes from thermophilic organisms relative to their homologues from mesophiles are known from earlier studies. We used cellulase C from Clostridium thermocellum to test whether thermostability can be increased by mutations designed using rules learned from thermophilic proteins. Cellulase C has a TIM barrel fold with an additional helical subdomain. We designed and produced a number of mutants with the aim to increase its thermostability. Five mutants were designed to create new electrostatic interactions. They all retained catalytic activity but exhibited decreased thermostability relative to the wild-type enzyme. Here, the stabilizing contributions are obviously smaller than the destabilization caused by the introduction of the new side chains. In another mutant, the small helical subdomain was deleted. This mutant lost activity but its melting point was only 3 degrees C lower than that of the wild-type enzyme, which suggests that the subdomain is an independent folding unit and is important for catalytic function. A double mutant was designed to introduce a new disulfide bridge into the enzyme. This mutant is active and has an increased stability (deltaT(m)=3 degrees C, delta(deltaG(u))=1.73 kcal/mol) relative to the wild-type enzyme. Reduction of the disulfide bridge results in destabilization and an altered thermal denaturation behavior. We conclude that rules learned from thermophilic proteins cannot be used in a straightforward way to increase the thermostability of a protein. Creating a crosslink such as a disulfide bond is a relatively sure-fire method but the stabilization may be smaller than calculated due to coupled destabilizing effects.

Hydrogenases are redox metalloenzymes in bacteria that catalyze the uptake or production of molecular hydrogen. Two homologous nickel-iron hydrogenases, HupSL and HydSL from the photosynthetic purple sulfur bacterium Thiocapsa roseopersicina, differ substantially in their thermal stabilities despite the high sequence similarity between them. The optimum temperature of HydSL activity is estimated to be at least 50 degrees C higher than that of HupSL. In this work, homology models of both proteins were constructed and analyzed for a number of structural properties. The comparison of the models reveals that the higher stability of HydSL can be attributed to increased inter-subunit electrostatic interactions: the homology models reliably predict that HydSL contains at least five more inter-subunit ion pairs than HupSL. The subunit interface of HydSL is more polar than that of HupSL, and it contains a few extra inter-subunit hydrogen bonds. A more optimized cavity system and amino acid replacements resulting in increased conformational rigidity may also contribute to the higher stability of HydSL. The results are in accord with the general observation that with increasing temperature, the role of electrostatic interactions in protein stability increases.

We have applied the TOUCHSTONE structure prediction algorithm that spans the range from homology modeling to ab initio folding to all protein targets in CASP5. Using our threading algorithm PROSPECTOR that does not utilize input from metaservers, one threads against a representative set of PDB templates. If a template is significantly hit, Generalized Comparative Modeling designed to span the range from closely to distantly related proteins from the template is done. This involves freezing the aligned regions and relaxing the remaining structure to accommodate insertions or deletions with respect to the template. For all targets, consensus predicted side chain contacts from at least weakly threading templates are pooled and incorporated into ab initio folding. Often, TOUCHSTONE performs well in the CM to FR categories, with PROSPECTOR showing significant ability to identify analogous templates. When ab initio folding is done, frequently the best models are closer to the native state than the initial template. Among the particularly good predictions are T0130 in the CM/FR category, T0138 in the FR(H) category, T0135 in the FR(A) category, T0170 in the FR/NF category and T0181 in the NF category. Improvements in the approach are needed in the FR/NF and NF categories. Nevertheless, TOUCHSTONE was one of the best performing algorithms over all categories in CASP5.

The structural clues of substrate recognition by calpain are incompletely understood. In this study, 106 cleavage sites in substrate proteins compiled from the literature have been analyzed to dissect the signal for calpain cleavage and also to enable the design of an ideal calpain substrate and interfere with calpain action via site-directed mutagenesis. In general, our data underline the importance of the primary structure of the substrate around the scissile bond in the recognition process. Significant amino acid preferences were found to extend over 11 residues around the scissile bond, from P(4) to P(7)'. In compliance with earlier data, preferred residues in the P(2) position are Leu, Thr, and Val, and in P(1) Lys, Tyr, and Arg. In position P(1) ', small hydrophilic residues, Ser and to a lesser extent Thr and Ala, occur most often. Pro dominates the region flanking the P(2)-P(1)' segment, i.e. positions P(3) and P(2)'-P(4)'; most notable is its occurrence 5.59 times above chance in P(3)'. Intriguingly, the segment C-terminal to the cleavage site resembles the consensus inhibitory region of calpastatin, the specific inhibitor of the enzyme. Further, the position of the scissile bond correlates with certain sequential attributes, such as secondary structure and PEST score, which, along with the amino acid preferences, suggests that calpain cleaves within rather disordered segments of proteins. The amino acid preferences were confirmed by site-directed mutagenesis of the autolysis sites of Drosophila calpain B; when amino acids at key positions were changed to less preferred ones, autolytic cleavage shifted to other, adjacent sites. Based on these preferences, a new fluorogenic calpain substrate, DABCYLTPLKSPPPSPR-EDANS, was designed and synthesized. In the case of micro- and m-calpain, this substrate is kinetically superior to commercially available ones, and it can be used for the in vivo assessment of the activity of these ubiquitous mammalian calpains.

Many essential cellular processes such as signal transduction, transport, cellular motion and most regulatory mechanisms are mediated by protein-protein interactions. In recent years, new experimental techniques have been developed to discover the protein-protein interaction networks of several organisms. However, the accuracy and coverage of these techniques have proven to be limited, and computational approaches remain essential both to assist in the design and validation of experimental studies and for the prediction of interaction partners and detailed structures of protein complexes. Here, we provide a critical overview of existing structure-independent and structure-based computational methods. Although these techniques have significantly advanced in the past few years, we find that most of them are still in their infancy. We also provide an overview of experimental techniques for the detection of protein-protein interactions. Although the developments are promising, false positive and false negative results are common, and reliable detection is possible only by taking a consensus of different experimental approaches. The shortcomings of experimental techniques affect both the further development and the fair evaluation of computational prediction methods. For an adequate comparative evaluation of prediction and high-throughput experimental methods, an appropriately large benchmark set of biophysically characterized protein complexes would be needed, but is sorely lacking.

Structural genomics projects as well as ab initio protein structure prediction methods provide structures of proteins with no sequence or fold similarity to proteins with known functions. These are often low-resolution structures that may only include the positions of C alpha atoms. We present a fast and efficient method to predict DNA-binding proteins from just the amino acid sequences and low-resolution, C alpha-only protein models. The method uses the relative proportions of certain amino acids in the protein sequence, the asymmetry of the spatial distribution of certain other amino acids as well as the dipole moment of the molecule. These quantities are used in a linear formula, with coefficients derived from logistic regression performed on a training set, and DNA-binding is predicted based on whether the result is above a certain threshold. We show that the method is insensitive to errors in the atomic coordinates and provides correct predictions even on inaccurate protein models. We demonstrate that the method is capable of predicting proteins with novel binding site motifs and structures solved in an unbound state. The accuracy of our method is close to another, published method that uses all-atom structures, time-consuming calculations and information on conserved residues.

Since Anfinsen's famous experiments in the 1960s, it has been known that the complex three-dimensional structure of protein molecules is encoded in their amino acid sequences, and the chains autonomously fold under proper conditions. Cracking this code, which is sometimes called "the second part of the genetic code," has been one of the greatest challenges of molecular biology. Although a full understanding of how proteins fold remains elusive, theoretical and experimental studies of protein folding have come a long way since Anfinsen's findings. In the living cell, folding occurs in a complex and crowded environment, often involving helper proteins, and in some cases it can go awry: the protein can misfold, aggregate, or form amyloid fibers. It is increasingly recognized that misfolded proteins and amyloid formation are the root cause of a number of serious illnesses including several neurodegenerative diseases. Therefore, the study of protein folding remains a key area of biomedical research.

The relationship between the thermal stability of proteins and rates of unfolding and refolding is still an open issue. The data are very scarce, especially for proteins with complex structure. Here, time-dependent denaturation-renaturation experiments on Thermus thermophilus, Escherichia coli, and Vibrio sp. I5 3-isopropylmalate dehydrogenases (IPMDHs) of different heat stabilities are presented. Unfolding, as monitored by several methods, occurs in a single first-order step with half-times of approximately 1 h, several minutes, and few seconds for the thermophilic, mesophilic, and psychrotrophic enzymes, respectively. The binding of Mn*IPM (the manganese complex of 3-isopropylmalate) markedly reduces the rates of unfolding; this effect is more prominent for the less stable enzyme variants. Refolding is a two-step or multistep first-order process involving an inactive intermediate(s). The restoration of the native structure and reactivation take place with a half-time of a few minutes for all three IPMDHs. Thus, the comparative experimental unfolding-refolding studies of the three IPMDHs with different thermostabilities have revealed a close relationship between thermostability and unfolding rate. Structural analysis has shown that the differences in the molecular contacts between selected nonconserved residues are responsible for the different rates of unfolding. On the other hand, the folding rates might be correlated with the absolute contact order, which does not significantly vary between IPMDHs with different thermostabilities. On the basis of our observations, folding rates appear to be dictated by global structural characteristics (such as native topology, i.e., contact order) rather than by thermodynamic stability.

In a paper titled "A topologically related singularity suggests a maximum preferred size for protein domains" (Zbilut et al., Proteins 2007;66:621-629), Zbilut et al. claim to have found a singularity in certain geometrical properties of protein structures, and suggest that this singularity may limit the maximum size of protein domains. They find further support for the singularity in their analysis of G-factors calculated by the PROCHECK program. Here, we show that the claimed singularity is a mathematical artifact with no physical meaning, and we reanalyze the G-factors to show that Zbilut et al.'s results are due to a single outlier in the data. Thus, the existence of an actual singularity in the topological properties of proteins is not supported by the findings of Zbilut et al.

The amino acid composition of intrinsically disordered proteins and protein segments characteristically differs from that of ordered proteins. This observation forms the basis of several disorder prediction methods. These, however, usually perform worse for smaller proteins (or segments) than for larger ones. We show that the regions of amino acid composition space corresponding to ordered and disordered proteins overlap with each other, and the extent of the overlap (the "twilight zone") is larger for short than for long chains. To explain this finding, we used two-dimensional lattice model proteins containing hydrophobic, polar, and charged monomers and revealed the relation among chain length, amino acid composition, and disorder. Because the number of chain configurations exponentially grows with chain length, a larger fraction of longer chains can reach a low-energy, ordered state than do shorter chains. The amount of information carried by the amino acid composition about whether a protein or segment is (dis)ordered grows with increasing chain length. Smaller proteins rely more on specific interactions for stability, which limits the possible accuracy of disorder prediction methods. For proteins in the "twilight zone", size can determine order, as illustrated by the example of two-state homodimers.

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from Thermotoga maritima (TmGAPDH) is a thermostable enzyme (Tm = 102 degrees C), which is fully active at temperatures near 80 degrees C but has very low activity at room temperature. In search for an explanation of this behavior, we measured the conformational flexibility of the protein by hydrogen-deuterium exchange and compared the results with those obtained with GAPDH from rabbit muscle (RmGAPDH). At room temperature, the conformational flexibility of TmGAPDH is much less than that of RmGAPDH, but increases with increasing temperature and becomes comparable to that of RmGAPDH near the physiological temperature of Thermotoga maritima. Using the available three-dimensional structures of the two enzymes, we compared the B factors that reflect the local mobility of protein atoms. The largest differences in B factors are seen in the coenzyme and NAD binding regions. The likely reason for the low activity of TmGAPDH at room temperature is that the motions required for enzyme functions are restricted. The findings support the idea of "corresponding states" which claims that over the time span of evolution, the overall conformational flexibility of proteins has been preserved at their corresponding physiological temperatures.

A possible approach to generate enzymes with an engineered temperature optimum is to create chimeras of homologous enzymes with different temperature optima. We tested this approach using two family-10 xylanases from Thermotoga maritima: the thermophilic xylanase A catalytic domain (TmxAcat, T(opt)=68 degrees C), and the hyperthermophilic xylanase B (TmxB, T(opt)=102 degrees C). Twenty-one different chimeric constructs were created by mimicking family shuffling in a rational manner. The measured temperature optima of the 16 enzymatically active chimeras do not monotonically increase with the percentage of residues coming from TmxB. Only four chimeras had a higher temperature optimum than TmxAcat, the most stable variant (T(opt)=80 degrees C) being the one in which both terminal segments came from TmxB. Further analysis suggests that the interaction between the N- and C-terminal segments has a disproportionately high contribution to the overall thermostability. The results may be generalizable to other enzymes where the N- and C-termini are in contact.

DNA-binding proteins (DBPs) participate in various crucial processes in the life-cycle of the cells, and the identification and characterization of these proteins is of great importance. We present here a random forests classifier for identifying DBPs among proteins with known 3D structures. First, clusters of evolutionarily conserved regions (patches) on the surface of proteins were detected using the PatchFinder algorithm; earlier studies showed that these regions are typically the functionally important regions of proteins. Next, we trained a classifier using features like the electrostatic potential, cluster-based amino acid conservation patterns and the secondary structure content of the patches, as well as features of the whole protein, including its dipole moment. Using 10-fold cross-validation on a dataset of 138 DBPs and 110 proteins that do not bind DNA, the classifier achieved a sensitivity and a specificity of 0.90, which is overall better than the performance of published methods. Furthermore, when we tested five different methods on 11 new DBPs that did not appear in the original dataset, only our method annotated all correctly. The resulting classifier was applied to a collection of 757 proteins of known structure and unknown function. Of these proteins, 218 were predicted to bind DNA, and we anticipate that some of them interact with DNA using new structural motifs. The use of complementary computational tools supports the notion that at least some of them do bind DNA.

Enzyme function depends on specific conformational motions. We show that the temperature dependence of enzyme kinetic parameters can provide insight into these functionally relevant motions. While investigating the catalytic properties of IPMDH from Escherichia coli, we found that its catalytic efficiency (k(cat)/K(M,IPM)) for the substrate IPM has an unusual temperature dependence, showing a local minimum at approximately 35 degrees C. In search of an explanation, we measured the individual constants k(cat) and K(M,IPM) as a function of temperature, and found that the van 't Hoff plot of K(M,IPM) shows sigmoid-like transition in the 20-40 degrees C temperature range. By means of various measurements including hydrogen-deuterium exchange and fluorescence resonance energy transfer, we showed that the conformational fluctuations, including hinge-bending domain motions increase more steeply with temperatures >30 degrees C. The thermodynamic parameters of ligand binding determined by isothermal titration calorimetry as a function of temperature were found to be strongly correlated to the conformational fluctuations of the enzyme. Because the binding of IPM is associated with a hinge-bending domain closure, the more intense hinge-bending fluctuations at higher temperatures increasingly interfere with IPM binding, thereby abruptly increasing its dissociation constant and leading to the observed unusual temperature dependence of the catalytic efficiency.

Galectins are proteins that regulate immune responses through the recognition of cell-surface glycans. We present evidence that 16 human galectin genes are expressed at the maternal-fetal interface and demonstrate that a cluster of 5 galectin genes on human chromosome 19 emerged during primate evolution as a result of duplication and rearrangement of genes and pseudogenes via a birth and death process primarily mediated by transposable long interspersed nuclear elements (LINEs). Genes in the cluster are found only in anthropoids, a group of primate species that differ from their strepsirrhine counterparts by having relatively large brains and long gestations. Three of the human cluster genes (LGALS13, -14, and -16) were found to be placenta-specific. Homology modeling revealed conserved three-dimensional structures of galectins in the human cluster; however, analyses of 24 newly derived and 69 publicly available sequences in 10 anthropoid species indicate functional diversification by evidence of positive selection and amino acid replacements in carbohydrate-recognition domains. Moreover, we demonstrate altered sugar-binding capacities of 6 recombinant galectins in the cluster. We show that human placenta-specific galectins are predominantly expressed by the syncytiotrophoblast, a primary site of metabolic exchange where, early during pregnancy, the fetus comes in contact with immune cells circulating in maternal blood. Because ex vivo functional assays demonstrate that placenta-specific galectins induce the apoptosis of T lymphocytes, we propose that these galectins reduce the danger of maternal immune attacks on the fetal semiallograft, presumably conferring additional immune tolerance mechanisms and in turn sustaining hemochorial placentation during the long gestation of anthropoid primates.

To annotate the biological function of a protein molecule, it is essential to have information on its 3D structure. Many successful methods for function prediction are based on determining structurally conserved regions because the functional residues are proved to be more conservative than others in protein evolution. Since the 3D conformation of a protein can be represented by a contact map graph, graph matching, algorithms are often employed to identify the conserved residues in weakly homologous protein pairs. However, the general graph matching algorithm is computationally expensive because graph similarity searching is essentially a NP-hard problem. Parallel implementations of the graph matching are often exploited to speed up the process. In this chapter, the authors review theoretical and computational approaches of graph theory and the recently developed graph matching algorithms for protein function prediction.

The post-genomic era has witnessed an explosion of protein sequences in the public databases; but this has not been complemented by the availability of genome-wide structure and function information, due to the technical difficulties and labor expenses incurred by existing experimental techniques. The rapid advancements in computer-based protein structure prediction methods have enabled automated and yet reliable methods for generating three-dimensional (3D) structural models of proteins. Genome-scale structure prediction experiments have been conducted by a number of groups, starting as early as in 1997, and some noteworthy efforts have been made using the MODELLER and ROSETTA methods. Along another line, TOUCHSTONE was used to predict the structures of all 85 small proteins in the Mycoplasma genitalium genome, which established template-refinement-based structure prediction as a practical approach for genome-scale experiments. This was followed by the development of Threading ASSEmbly Refinement (TASSER) and Iterative Threading ASSEmbly Refinement (I-TASSER) algorithms which use a combination of various approaches for threading, fragment assembly, ab initio loop modeling, and structural refinement to predict the structures. A successful structural prediction for all medium-sized open reading frames (ORFs) in the Escherichia coli genome was demonstrated by this method, achieving high-accuracy models for 920 out of 1,360 proteins. G protein-coupled receptors (GPCRs) are an extremely important class of membrane proteins for which only very few structures are available in the Protein Data Bank (PDB). TASSER was used to predict the structures of all 907 putative GPCRs in the human genome, and the high accuracy confirmed by newly solved GPCR structures and recent blind tests have demonstrated the usefulness and robustness of the TASSER/I-TASSER models for the functional annotation of GPCRs. Recently, the I-TASSER protein structure prediction method has been used as a basis for functional annotation of protein sequences. The increasing popularity and need for such automated structure and function prediction algorithms can be judged by the fact that the I-TASSER server has generated structure predictions for 35,000 proteins submitted by more than 8,000 users from 86 countries in the last 24 months. The success of these modeling experiments demonstrates significant new progress in high-throughput and genome-wide protein structure prediction.

The iDBPs server uses the three-dimensional (3D) structure of a query protein to predict whether it binds DNA. First, the algorithm predicts the functional region of the protein based on its evolutionary profile; the assumption is that large clusters of conserved residues are good markers of functional regions. Next, various characteristics of the predicted functional region as well as global features of the protein are calculated, such as the average surface electrostatic potential, the dipole moment and cluster-based amino acid conservation patterns. Finally, a random forests classifier is used to predict whether the query protein is likely to bind DNA and to estimate the prediction confidence. We have trained and tested the classifier on various datasets and shown that it outperformed related methods. On a dataset that reflects the fraction of DNA binding proteins (DBPs) in a proteome, the area under the ROC curve was 0.90. The application of the server to an updated version of the N-Func database, which contains proteins of unknown function with solved 3D-structure, suggested new putative DBPs for experimental studies.

Although residue-residue contact maps dictate the topology of proteins, sequence-based ab initio contact predictions have been found little use in actual structure prediction due to the low accuracy. We developed a composite set of nine SVM-based contact predictors that are used in I-TASSER simulation in combination with sparse template contact restraints. When testing the strategy on 273 nonhomologous targets, remarkable improvements of I-TASSER models were observed for both easy and hard targets, with p value by Student's t test<0.00001 and 0.001, respectively. In several cases, template modeling score increases by >30%, which essentially converts "nonfoldable" targets into "foldable" ones. In CASP9, I-TASSER employed ab initio contact predictions, and generated models for 26 FM targets with a GDT-score 16% and 44% higher than the second and third best servers from other groups, respectively. These findings demonstrate a new avenue to improve the accuracy of protein structure prediction especially for free-modeling targets.

BACKGROUND: Placental Protein 13 (PP13), an early biomarker of preeclampsia, is a placenta-specific galectin that binds beta-galactosides, building-blocks of ABO blood-group antigens, possibly affecting its bioavailability in blood. METHODS AND FINDINGS: We studied PP13-binding to erythrocytes, maternal blood-group effect on serum PP13 and its performance as a predictor of preeclampsia and intrauterine growth restriction (IUGR). Datasets of maternal serum PP13 in Caucasian (n.=.1078) and Hispanic (n.=.242) women were analyzed according to blood groups. In vivo, in vitro and in silico PP13-binding to ABO blood-group antigens and erythrocytes were studied by PP13-immunostainings of placental tissue-microarrays, flow-cytometry of erythrocyte-bound PP13, and model-building of PP13--blood-group H antigen complex, respectively. Women with blood group AB had the lowest serum PP13 in the first trimester, while those with blood group B had the highest PP13 throughout pregnancy. In accordance, PP13-binding was the strongest to blood-group AB erythrocytes and weakest to blood-group B erythrocytes. PP13-staining of maternal and fetal erythrocytes was revealed, and a plausible molecular model of PP13 complexed with blood-group H antigen was built. Adjustment of PP13 MoMs to maternal ABO blood group improved the prediction accuracy of first trimester maternal serum PP13 MoMs for preeclampsia and IUGR. CONCLUSIONS: ABO blood group can alter PP13-bioavailability in blood, and it may also be a key determinant for other lectins' bioavailability in the circulation. The adjustment of PP13 MoMs to ABO blood group improves the predictive accuracy of this test.

Multidomain proteins form in evolution through the concatenation of domains, but structural domains may comprise multiple segments of the chain. In this work, we demonstrate that new multidomain architectures can evolve by an apparent three-dimensional swap of segments between structurally similar domains within a single-chain monomer. By a comprehensive structural search of the current Protein Data Bank (PDB), we identified 32 well-defined segment-swapped proteins (SSPs) belonging to 18 structural families. Nearly 13% of all multidomain proteins in the PDB may have a segment-swapped evolutionary precursor as estimated by more permissive searching criteria. The formation of SSPs can be explained by two principal evolutionary mechanisms: (i) domain swapping and fusion (DSF) and (ii) circular permutation (CP). By large-scale comparative analyses using structural alignment and hidden Markov model methods, it was found that the majority of SSPs have evolved via the DSF mechanism, and a much smaller fraction, via CP. Functional analyses further revealed that segment swapping, which results in two linkers connecting the domains, may impart directed flexibility to multidomain proteins and contributes to the development of new functions. Thus, inter-domain segment swapping represents a novel general mechanism by which new protein folds and multidomain architectures arise in evolution, and SSPs have structural and functional properties that make them worth defining as a separate group.

"Pull moves" is a popular move set for lattice polymer model simulations. We show that the proof given for its reversibility earlier is flawed, and some moves are irreversible, which leads to biases in the parameters estimated from the simulations. We show how to make the move set fully reversible.

The numerous discovered cases of domesticated transposable element (TE) proteins led to the recognition that TEs are a significant source of evolutionary innovation. However, much less is known about the reverse process, whether and to what degree the evolution of TEs is influenced by the genome of their hosts. We addressed this issue by searching for cases of incorporation of host genes into the sequence of TEs and examined the systems-level properties of these genes using the Saccharomyces cerevisiae and Drosophila melanogaster genomes. We identified 51 cases where the evolutionary scenario was the incorporation of a host gene fragment into a TE consensus sequence, and we show that both the yeast and fly homologues of the incorporated protein sequences have central positions in the cellular networks. An analysis of selective pressure (Ka/Ks ratio) detected significant selection in 37% of the cases. Recent research on retrovirus-host interactions shows that virus proteins preferentially target hubs of the host interaction networks enabling them to take over the host cell using only a few proteins. We propose that TEs face a similar evolutionary pressure to evolve proteins with high interacting capacities and take some of the necessary protein domains directly from their hosts.

Allosteric drugs bind to sites which are usually less conserved evolutionarily as compared to orthosteric sites. As such, they can discriminate between closely related proteins, have fewer side effects, and a consequent lower concentration can convey a lesser likelihood of receptor desensitization. However, an allosteric mode of action may also make the results of preclinical and animal experiments less predictive. The sensitivity of the allosteric consequences to the environment further increases the importance of accounting for patient population diversity. Even subtle differences in protein sequence, in cellular metabolic states or in target tissues, can result in different outcomes. This mini-hot-topic issue of CTMC showcases some successes and challenges of allosteric drug development through the examples of seventransmembrane (GPCR), AMPA, NMDA and metabotropic glutamate receptors, as well as the morpheein model of allosterism involved in inherent metabolic errors. Finally, the development of allo-network drugs, which are allosteric drugs acting indirectly on the neighborhood of the pharmacological target in protein-protein interaction or signaling networks, is described.

Allosteric drugs are usually more specific and have fewer side effects than orthosteric drugs targeting the same protein. Here, we overview the current knowledge on allosteric signal transmission from the network point of view, and show that most intra-protein conformational changes may be dynamically transmitted across protein-protein interaction and signaling networks of the cell. Allo-network drugs influence the pharmacological target protein indirectly using specific inter-protein network pathways. We show that allo-network drugs may have a higher efficiency to change the networks of human cells than those of other organisms, and can be designed to have specific effects on cells in a diseased state. Finally, we summarize possible methods to identify allo-network drug targets and sites, which may develop to a promising new area of systems-based drug design.

Despite the considerable amount of research on transposable elements, no large-scale structural analyses of the TE proteome have been performed so far. We predicted the structures of hundreds of proteins from a representative set of DNA and LINE transposable elements and used the obtained structural data to provide the first general structural characterization of TE proteins and to estimate the frequency of TE domestication and horizontal transfer events. We show that 1) ORF1 and Gag proteins of retrotransposons contain high amounts of structural disorder; thus, despite their very low conservation, the presence of disordered regions and probably their chaperone function is conserved. 2) The distribution of SCOP classes in DNA transposons and LINEs indicates that the proteins of DNA transposons are more ancient, containing folds that already existed when the first cellular organisms appeared. 3) DNA transposon proteins have lower contact order than randomly selected reference proteins, indicating rapid folding, most likely to avoid protein aggregation. 4) Structure-based searches for TE homologs indicate that the overall frequency of TE domestication events is low, whereas we found a relatively high number of cases where horizontal transfer, frequently involving parasites, is the most likely explanation for the observed homology.

To annotate the biological function of a protein molecule, it is essential to have information on its 3D structure. Many successful methods for function prediction are based on determining structurally conserved regions because the functional residues are proved to be more conservative than others in protein evolution. Since the 3D conformation of a protein can be represented by a contact map graph, graph matching, algorithms are often employed to identify the conserved residues in weakly homologous protein pairs. However, the general graph matching algorithm is computationally expensive because graph similarity searching is essentially a NP-hard problem. Parallel implementations of the graph matching are often exploited to speed up the process. In this chapter, the authors review theoretical and computational approaches of graph theory and the recently developed graph matching algorithms for protein function prediction.

The structure of protein-protein complexes can be constructed by using the known structure of other protein complexes as a template. The complex structure templates are generally detected either by homology-based sequence alignments or, given the structure of monomer components, by structure-based comparisons. Critical improvements have been made in recent years by utilizing interface recognition and by recombining monomer and complex template libraries. Encouraging progress has also been witnessed in genome-wide applications of template-based modeling, with modeling accuracy comparable to high-throughput experimental data. Nevertheless, bottlenecks exist due to the incompleteness of the protein-protein complex structure library and the lack of methods for distant homologous template identification and full-length complex structure refinement.

Galectins are glycan-binding proteins that regulate innate and adaptive immune responses, and some confer maternal-fetal immune tolerance in eutherian mammals. A chromosome 19 cluster of galectins has emerged in anthropoid primates, species with deep placentation and long gestation. Three of the five human cluster galectins are solely expressed in the placenta, where they may confer additional immunoregulatory functions to enable deep placentation. One of these is galectin-13, also known as Placental Protein 13 (PP13). It has a "jelly-roll" fold, carbohydrate-recognition domain and sugar-binding preference resembling other mammalian galectins. PP13 is predominantly expressed by the syncytiotrophoblast and released from the placenta into the maternal circulation. Its ability to induce apoptosis of activated T cells in vitro, and to divert and kill T cells as well as macrophages in the maternal decidua in situ, suggests important immune functions. Indeed, mutations in the promoter and an exon of LGALS13 presumably leading to altered or non-functional protein expression are associated with a higher frequency of preeclampsia and other obstetrical syndromes, which involve immune dysregulation. Moreover, decreased placental expression of PP13 and its low concentrations in first trimester maternal sera are associated with elevated risk of preeclampsia. Indeed, PP13 turned to be a good early biomarker to assess maternal risk for the subsequent development of pregnancy complications caused by impaired placentation. Due to the ischemic placental stress in preterm preeclampsia, there is increased trophoblastic shedding of PP13 immunopositive microvesicles starting in the second trimester, which leads to high maternal blood PP13 concentrations. Our meta-analysis suggests that this phenomenon may enable the potential use of PP13 in directing patient management near to or at the time of delivery. Recent findings on the beneficial effects of PP13 on decreasing blood pressure due to vasodilatation in pregnant animals suggest its therapeutic potential in preeclampsia.

3-Isopropylmalate dehydrogenase (IPMDH) from Mycobacterium tuberculosis (Mtb) may be a target for specific drugs against this pathogenic bacterium. We have expressed and purified Mtb IPMDH and determined its physical-chemical and enzymological properties. Size-exclusion chromatography and dynamic light scattering measurements (DLS) suggest a tetrameric structure for Mtb IPMDH, in contrast to the dimeric structure of most IPMDHs. The kinetic properties (kcat and Km values) of Mtb IPMDH and the pH-dependence of kcat are very similar to both Escherichia coli (Ec) and Thermus thermophilus (Tt) IPMDHs. The stability of Mtb IPMDH in 8 M urea is close to that of the mesophilic counterpart, Ec IPMDH, both of them being much less stable than the thermophilic (Tt) enzyme. Two known IPMDH inhibitors, O-methyl oxalohydroxamate and 3-methylmercaptomalate, have been synthesised. Their inhibitory effects were found to be independent of the origin of IPMDHs. Thus, experiments with either Ec or Tt IPMDH would be equally relevant for designing specific inhibitory drugs against Mtb IPMDH.

Transposases are important tools in genome engineering, and there is considerable interest in engineering more efficient ones. Here we seek to understand the factors determining their activity using the Sleeping Beauty transposase. Recent work suggests that protein co-evolutionary information can be used to classify groups of physically connected, co-evolving residues into elements called ‘sectors’, which have proven useful for understanding the folding, allosteric interactions, and enzymatic activity of proteins. Using extensive mutagenesis data, protein modeling and analysis of folding energies, we show that 1) The Sleeping Beauty transposase contains two sectors, which span across conserved domains, and are enriched in DNA-binding residues, indicating that the DNA binding and endonuclease functions of the transposase coevolve; 2) Sector residues are highly sensitive to mutations, and most mutations of these residues strongly reduce transposition rate; 3) Mutations with a strong effect on free energy of folding in the DDE domain of the transposase significantly reduce transposition rate. 4) Mutations that influence DNA and protein-protein interactions generally reduce transposition rate, although most hyperactive mutants are also located on the protein surface, including residues with protein-protein interactions. This suggests that hyperactivity results from the modification of protein interactions, rather than the stabilization of protein fold.

In an earlier study, we showed that two-domain segment-swapped proteins can evolve by domain swapping and fusion, resulting in a protein with two linkers connecting its domains. We proposed that a potential evolutionary advantage of this topology may be the restriction of interdomain motions, which may facilitate domain closure by a hinge-like movement, crucial for the function of many enzymes. Here, we test this hypothesis computationally on uroporphyrinogen III synthase, a two-domain segment-swapped enzyme essential in porphyrin metabolism. To compare the interdomain flexibility between the wild-type, segment-swapped enzyme (having two interdomain linkers) and circular permutants of the same enzyme having only one interdomain linker, we performed geometric and molecular dynamics simulations for these species in their ligand-free and ligand-bound forms. We find that in the ligand-free form, interdomain motions in the wild-type enzyme are significantly more restricted than they would be with only one interdomain linker, while the flexibility difference is negligible in the ligand-bound form. We also estimated the entropy costs of ligand binding associated with the interdomain motions, and find that the change in domain connectivity due to segment swapping results in a reduction of this entropy cost, corresponding to ∼20% of the total ligand binding free energy. In addition, the restriction of interdomain motions may also help the functional domain-closure motion required for catalysis. This suggests that the evolution of the segment-swapped topology facilitated the evolution of enzyme function for this protein by influencing its dynamic properties. This article is protected by copyright. All rights reserved.

We present a novel, conceptually simple approach to calculate the configurational entropy difference between two conformational ensembles of a molecular system. The method estimates the full-dimensional probability density function of the system by a Gaussian mixture, using an efficient greedy learning algorithm with a cross-validation based stopping criterion. Evaluating the method on conformational ensembles corresponding to substates of five small peptide systems, excellent agreement is found with the exact entropy differences obtained from a full enumeration of conformations. Compared with the quasiharmonic method and two other, more recently developed methods, the Gaussian mixture method yields more accurate results at smaller sample sizes. We illustrate the power of the method by calculating the backbone torsion angle entropy difference between disulfide-bonded and non-disulfide-bonded states of tachyplesin, a 17-residue antimicrobial peptide, and between two substates in the native ensemble of the 58-residue bovine pancreatic trypsin inhibitor. The program is available at https://gmentropy.szialab.org.

Serine proteases (SPs) are typically synthesized as precursors, termed proenzymes or zymogens, and the fully active form is produced via limited proteolysis by another protease or by autoactivation. The lectin pathway of the complement system is initiated by mannose-binding lectin (MBL)-associated SPs (MASP)-1, and MASP-2, which are known to be present as proenzymes in blood. The third SP of the lectin pathway, MASP-3, was recently shown to be the major activator, and the exclusive "resting blood" activator of profactor D, producing factor D, the initiator protease of the alternative pathway. Because only activated MASP-3 is capable of carrying out this cleavage, it was presumed that a significant fraction of MASP-3 must be present in the active form in resting blood. Here, we aimed to detect active MASP-3 in the blood by a more direct technique and to quantitate the active to zymogen ratio. First, MASPs were partially purified (enriched) from human plasma samples by affinity chromatography using immobilized MBL in the presence of inhibitors. Using this MASP pool, only the zymogen form of MASP-1 was detected by Western blot, whereas over 70% MASP-3 was in an activated form in the same samples. Furthermore, the active to zymogen ratio of MASP-3 showed little individual variation. It is enigmatic how MASP-3, which is not able to autoactivate, is present mostly as an active enzyme, whereas MASP-1, which has a potent autoactivation capability, is predominantly proenzymic in resting blood. In an attempt to explain this phenomenon, we modeled the basal level fluid-phase activation of lectin pathway proteases and their subsequent inactivation by C1 inhibitor and antithrombin using available and newly determined kinetic constants. The model can explain extensive MASP-3 activation only if we assume efficient intracomplex activation of MASP-3 by zymogen MASP-1. On the other hand, the model is in good agreement with the fact that MASP-1 and -2 are predominantly proenzymic and some of them is present in the form of inactive serpin-protease complexes. As an alternative hypothesis, MASP-3 activation by proprotein convertases is also discussed.

The complex effects of estradiol on non-reproductive tissues/cells, including lymphoid tissues and immunocytes, have increasingly been explored. However, the role of sex hormone binding globulin (SHBG) in the regulation of these genomic and non-genomic actions of estradiol is controversial. Moreover, the expression of SHBG and its internalization by potential receptors, as well as the influence of SHBG on estradiol uptake and signaling in lymphocytes has remained unexplored. Here, we found that human and mouse T cells expressed SHBG intrinsically. In addition, B lymphoid cell lines as well as both primary B and T lymphocytes bound and internalized external SHBG, and the amount of plasma membrane-bound SHBG decreased in B cells of pregnant compared to non-pregnant women. As potential mediators of this process, SHBG receptor candidates expressed by lymphocytes were identified in silico, including estrogen receptor (ER) alpha. Furthermore, cell surface-bound SHBG was detected in close proximity to membrane ERs while highly colocalizing with lipid rafts. The SHBG-membrane ER interaction was found functional since SHBG promoted estradiol uptake by lymphocytes and subsequently influenced Erk1/2 phosphorylation. In conclusion, the SHBG-SHBG receptor-membrane ER complex participates in the rapid estradiol signaling in lymphocytes, and this pathway may be altered in B cells in pregnant women.

The tumour suppressor protein RASSF1A is phosphorylated by Aurora A kinase, thereby impairing its tumour suppressor function. Consequently, inhibiting the interaction between Aurora A and RASSF1A may be used for anti-tumour therapy. We used recombinant variants of RASSF1A to map the sites of interaction with Aurora A. The phosphorylation kinetics of three truncated RASSF1A variants has been analysed. Compared to the RASSF1A form lacking the 120 residue long N-terminal part, the Km value of the phosphorylation is increased from 10 to 45 μM upon additional deletion of the C-terminal SARAH domain. On the other hand, deletion of the flexible loop (Δ177-197) that precedes the phosphorylation site/s (T202/S203) results in a reduction of the kcat value from about 40 to 7 min-1. Direct physical interaction between the isolated SARAH domain and Aurora A was revealed by SPR. These data demonstrate that the SARAH domain of RASSF1A is involved in the binding to Aurora A kinase. Structural modelling confirms that a novel complex is feasible between the SARAH domain and the kinase domain of Aurora A. In addition, a regulatory role of the loop in the catalytic phosphorylation reaction has been demonstrated both experimentally and by structural modelling.

Gene expression studies of molar pregnancy have been limited to a small number of candidate loci. We analyzed high-dimensional RNA and protein data to characterize molecular features of complete hydatidiform moles (CHMs) and corresponding pathologic pathways. CHMs and first trimester placentas were collected, histopathologically examined, then flash-frozen or paraffin-embedded. Frozen CHMs and control placentas were subjected to RNA-Seq, with resulting data and published placental RNA-Seq data subjected to bioinformatics analyses. Paraffin-embedded tissues from CHMs and control placentas were used for tissue microarray (TMA) construction, immunohistochemistry, and immunoscoring for galectin-14. Of the 14,022 protein-coding genes expressed in all samples, 3,729 were differentially expressed (DE) in CHMs, of which 72% were up-regulated. DE genes were enriched in placenta-specific genes (OR = 1.88, p = 0.0001), of which 79% were down-regulated, imprinted genes (OR = 2.38, p = 1.54 × 10−6), and immune genes (OR = 1.82, p = 7.34 × 10−18), of which 73% were up-regulated. DNA methylation-related enzymes and histone demethylases were dysregulated. “Cytokine–cytokine receptor interaction” was the most impacted of 38 dysregulated pathways, among which 17 were immune-related pathways. TMA-based immunoscoring validated the lower expression of galectin-14 in CHM. In conclusion, placental functions were down-regulated, imprinted gene expression was altered, and immune pathways were activated, indicating complex dysregulation of placental developmental and immune processes in CHMs.

The human placenta maintains pregnancy and supports the developing fetus by providing nutrition, gas-waste exchange, hormonal regulation, and an immunological barrier from the maternal immune system. The villous syncytiotrophoblast carries most of these functions and provides the interface between the maternal and fetal circulatory systems. The syncytiotrophoblast is generated by the biochemical and morphological differentiation of underlying cytotrophoblast progenitor cells. The dysfunction of the villous trophoblast development is implicated in placenta-mediated pregnancy complications. Herein, we describe gene modules and clusters involved in the dynamic differentiation of villous cytotrophoblasts into the syncytiotrophoblast. During this process, the immune defense functions are first established, followed by structural and metabolic changes, and then by peptide hormone synthesis. We describe key transcription regulatory molecules that regulate gene modules involved in placental functions. Based on transcriptomic evidence, we infer how villous trophoblast differentiation and functions are dysregulated in preterm preeclampsia, a life-threatening placenta-mediated obstetrical syndrome for the mother and fetus. In the conclusion, we uncover the blueprint for villous trophoblast development and its impairment in preterm preeclampsia, which may aid in the future development of non-invasive biomarkers for placental functions and early identification of women at risk for preterm preeclampsia as well as other placenta-mediated pregnancy complications.

Various pregnancy complications, such as severe forms of preeclampsia or intrauterine growth restriction, are thought to arise from failures in the differentiation of human placental trophoblasts. Progenitors of the latter either develop into invasive extravillous trophoblasts, remodeling the uterine vasculature, or fuse into multinuclear syncytiotrophoblasts transporting oxygen and nutrients to the growing fetus. However, key regulatory factors controlling trophoblast self-renewal and differentiation have been poorly elucidated. Using primary cells, three-dimensional organoids, and CRISPR-Cas9 genome-edited JEG-3 clones, we herein show that YAP, the transcriptional coactivator of the Hippo signaling pathway, promotes maintenance of cytotrophoblast progenitors by different genomic mechanisms. Genetic or chemical manipulation of YAP in these cellular models revealed that it stimulates proliferation and expression of cell cycle regulators and stemness-associated genes, but inhibits cell fusion and production of syncytiotrophoblast (STB)-specific proteins, such as hCG and GDF15. Genome-wide comparisons of primary villous cytotrophoblasts overexpressing constitutively active YAP-5SA with YAP KO cells and syncytializing trophoblasts revealed common target genes involved in trophoblast stemness and differentiation. ChIP-qPCR unraveled that YAP-5SA overexpression increased binding of YAP-TEAD4 complexes to promoters of proliferation-associated genes such as CCNA and CDK6 Moreover, repressive YAP-TEAD4 complexes containing the histone methyltransferase EZH2 were detected in the genomic regions of the STB-specific CGB5 and CGB7 genes. In summary, YAP plays a pivotal role in the maintenance of the human placental trophoblast epithelium. Besides activating stemness factors, it also directly represses genes promoting trophoblast cell fusion.

Introduction: Placental Protein 1 (PP1), PP8, and PP22 were isolated from the placenta. Herein, we aimed to identify PP1, PP8, and PP22 proteins and their placental and trophoblastic expression patterns to reveal potential involvement in pregnancy complications. Methods: We analyzed PP1, PP8, and PP22 proteins with LC-MS. We compared the placental behaviors of PP1, PP8, and PP22 to the predominantly placenta-expressed PP5/TFPI-2. Placenta-specificity scores were generated from microarray data. Trophoblasts were isolated from healthy placentas and differentiated; total RNA was isolated and subjected to microarray analysis. We assigned the placentas to the following groups: preterm controls, early-onset preeclampsia, early-onset preeclampsia with HELLP syndrome, term controls, and late-onset preeclampsia. After histopathologic examination, placentas were used for tissue microarray construction, immunostaining with anti-PP1, anti-PP5, anti-PP8, or anti-PP22 antibodies, and immunoscoring. Results: PP1, PP8, and PP22 were identified as 'nicotinate-nucleotide pyrophosphorylase', 'serpin B6', and 'protein disulfide-isomerase', respectively. Genes encoding PP1, PP8, and PP22 are not predominantly placenta-expressed, in contrast with PP5. PP1, PP8, and PP22 mRNA expression levels did not increase during trophoblast differentiation, in contrast with PP5. PP1, PP8, and PP22 immunostaining were detected primarily in trophoblasts, while PP5 expression was restricted to the syncytiotrophoblast. The PP1 immunoscore was higher in late-onset preeclampsia, while the PP5 immunoscore was higher in early-onset preeclampsia. Discussion: PP1, PP8, and PP22 are expressed primarily in trophoblasts but do not have trophoblast-specific regulation or functions. The distinct dysregulation of PP1 and PP5 expression in either late-onset or early-onset preeclampsia reflects different pathophysiological pathways in these preeclampsia subsets.

Zinc finger protein 554 (ZNF554), a member of the Krüppel-associated box domain zinc finger protein subfamily, is predominantly expressed in the brain and placenta in humans. Recently, we unveiled that ZNF554 regulates trophoblast invasion during placentation and its decreased expression leads to the early pathogenesis of preeclampsia. Since ZNF proteins are immensely implicated in the development of several tumors including malignant tumors of the brain, here we explored the pathological role of ZNF554 in gliomas. We examined the expression of ZNF554 at mRNA and protein levels in normal brain and gliomas, and then we searched for genome-wide transcriptomic changes in U87 glioblastoma cells transiently overexpressing ZNF554. Immunohistochemistry of brain tissues in our cohort (n = 62) and analysis of large TCGA RNA-Seq data (n = 687) of control, oligodendroglioma, and astrocytoma tissues both revealed decreased expression of ZNF554 towards higher glioma grades. Furthermore, low ZNF554 expression was associated with shorter survival of grade III and IV astrocytoma patients. Overexpression of ZNF554 in U87 cells resulted in differential expression, mostly downregulation of 899 genes. The "PI3K-Akt signaling pathway", known to be activated during glioma development, was the most impacted among 116 dysregulated pathways. Most affected pathways were cancer-related and/or immune-related. Congruently, cell proliferation was decreased and cell cycle was arrested in ZNF554-transfected glioma cells. These data collectively suggest that ZNF554 is a potential tumor suppressor and its decreased expression may lead to the loss of oncogene suppression, activation of tumor pathways, and shorter survival of patients with malignant glioma.

Rho-associated protein kinase 2 (ROCK2) is a membrane-anchored, long, flexible, multidomain, multifunctional protein. Its functions can be divided into two categories: membrane-proximal and membrane-distal. A recent study concluded that membrane-distal functions require the fully extended conformation, and this conclusion was supported by electron microscopy. The present solution small-angle X-ray scattering (SAXS) study revealed that ROCK2 population is a dynamic mixture of folded and partially extended conformers. Binding of RhoA to the coiled-coil domain shifts the equilibrium towards the partially extended state. Enzyme activity measurements suggest that the binding of natural protein substrates to the kinase domain breaks up the interaction between the N-terminal kinase and C-terminal regulatory domains, but smaller substrate analogues do not. The present study reveals the dynamic behaviour of this long, dimeric molecule in solution, and our structural model provides a mechanistic explanation for a set of membrane-proximal functions while allowing for the existence of an extended conformation in the case of membrane-distal functions.

Identification of pregnancies at risk of preterm birth (PTB), the leading cause of newborn deaths, remains challenging given the syndromic nature of the disease. We report a longitudinal multi-omics study coupled with a DREAM challenge to develop predictive models of PTB. The findings indicate that whole-blood gene expression predicts ultrasound-based gestational ages in normal and complicated pregnancies (r = 0.83) and, using data collected before 37 weeks of gestation, also predicts the delivery date in both normal pregnancies (r = 0.86) and those with spontaneous preterm birth (r = 0.75). Based on samples collected before 33 weeks in asymptomatic women, our analysis suggests that expression changes preceding preterm prelabor rupture of the membranes are consistent across time points and cohorts and involve leukocyte-mediated immunity. Models built from plasma proteomic data predict spontaneous preterm delivery with intact membranes with higher accuracy and earlier in pregnancy than transcriptomic models (AUROC = 0.76 versus AUROC = 0.6 at 27–33 weeks of gestation).

Gestational trophoblastic diseases (GTDs) have not been investigated for their epigenetic marks and consequent transcriptomic changes. Here, we analyzed genome-wide DNA methylation and transcriptome data to reveal the epigenetic basis of disease pathways that may lead to benign or malignant GTDs. RNA-Seq, mRNA microarray, and Human Methylation 450 BeadChip data from complete moles and choriocarcinoma cells were bioinformatically analyzed. Paraffin-embedded tissues from complete moles and control placentas were used for tissue microarray construction, DNMT3B immunostaining and immunoscoring. We found that DNA methylation increases with disease severity in GTDs. Differentially expressed genes are mainly upregulated in moles while predominantly downregulated in choriocarcinoma. DNA methylation principally influences the gene expression of villous trophoblast differentiation-related or predominantly placenta-expressed genes in moles and choriocarcinoma cells. Affected genes in these subsets shared focal adhesion and actin cytoskeleton pathways in moles and choriocarcinoma. In moles, cell cycle and differentiation regulatory pathways, essential for trophoblast/placental development, were enriched. In choriocarcinoma cells, hormone biosynthetic, extracellular matrix-related, hypoxic gene regulatory, and differentiation-related signaling pathways were enriched. In moles, we found slight upregulation of DNMT3B protein, a developmentally important de novo DNA methylase, which is strongly overexpressed in choriocarcinoma cells that may partly be responsible for the large DNA methylation differences. Our findings provide new insights into the shared and disparate molecular pathways of disease in GTDs and may help in designing new diagnostic and therapeutic tools.

Preeclampsia is a syndromic disease of the mother, fetus, and placenta. The main limitation in early and accurate diagnosis of preeclampsia is rooted in the heterogeneity of this syndrome as reflected by diverse molecular pathways, symptoms and clinical outcomes. Gaps in our knowledge preclude successful early diagnosis, personalized treatment and prevention. The advent of “omics” technologies and systems biology approaches enable addressing this problem by identifying the molecular pathways associated with the underlying mechanisms and clinical phenotypes of preeclampsia. Here, we provide a brief overview on how the field has progressed, focusing on studies utilizing state-of-the-art transcriptomics and proteomics methods. Moreover, we summarize our systems biology studies involving maternal blood proteomics and placental transcriptomics, which identified early maternal and placental disease pathways, and showed that their interaction influences the clinical presentation of preeclampsia. We also present an analysis of maternal blood proteomics data which revealed distinct molecular subclasses of preeclampsia and their molecular mechanisms. Maternal and placental disease pathways behind these subclasses are similar to those recently reported in studies on the placental transcriptome. These findings may promote the development of novel diagnostic tools for the distinct subtypes of preeclampsia syndrome, enabling early detection and personalized follow-up and tailored care of patients.

Objectives: The heterogeneous nature of preeclampsia is a major obstacle to early screening and prevention, and a molecular taxonomy of disease is needed. We have previously identified four subclasses of preeclampsia based on first-trimester plasma proteomic profiles. Herein, we expanded this approach by using a more comprehensive panel of proteins profiled in longitudinal samples. Methods: Proteomic data collected longitudinally from plasma samples of women who developed preeclampsia (n=109) and of controls (n=90) were available from our previous report on 1,125 proteins. Consensus clustering was performed to identify subgroups of patients with preeclampsia based on data from five gestational-age intervals by using select interval-specific features. Demographic, clinical, and proteomic differences among clusters were determined. Differentially abundant proteins were used to identify cluster-specific perturbed KEGG pathways. Results: Four molecular clusters with different clinical phenotypes were discovered by longitudinal proteomic profiling. Cluster 1 involves metabolic and prothrombotic changes with high rates of early-onset preeclampsia and small-for-gestational-age neonates; Cluster 2 includes maternal anti-fetal rejection mechanisms and recurrent preeclampsia cases; Cluster 3 is associated with extracellular matrix regulation and comprises cases of mostly mild, late-onset preeclampsia; and Cluster 4 is characterized by angiogenic imbalance and a high prevalence of early-onset disease. Conclusions: This study is an independent validation and further refining of molecular subclasses of preeclampsia identified by a different proteomic platform and study population. The results lay the groundwork for novel diagnostic and personalized tools of prevention.

The amyloidogenic processing of APP depends on two events: its phosphorylation by ROCK2 (at Thr654) and the phosphorylation of the APP-cleaving enzyme BACE1 (at Ser498). However, the mechanisms and structural details of APP-ROCK2 and BACE1-ROCK2 binding are unknown. Using direct physical methods in combination with an in silico approach, we found that BACE1 binds into the substrate-binding groove of ROCK2 with a low affinity (Kd = 18 µM), while no binding of APP to ROCK2 alone could be detected. On the other hand, a strong association (Kd = 3.5 nM) of APP to the weak ROCK2-BACE1 complex was observed, although no stable ternary complex was detected, i.e., BACE1 was displaced by APP. We constructed a sequential functional model: (1) BACE1 weakly binds to ROCK2 and induces an allosteric conformational change in ROCK2; (2) APP strongly binds to the ROCK2-BACE1 complex, and BACE1 is released; and (3) ROCK2 phosphorylates APP at Thr654 (leading to a longer stay in the early endosome during APP processing). Direct fluorescence titration experiments showed that the APP646–664 or APP665–695 fragments did not bind separately to the ROCK2-BACE1 complex. Based on these observations, we conclude that two binding sites are involved in the ROCK2-APP interaction: (1) the substrate-binding groove, where the APP646–664 sequence containing Thr654 sits and (2) the allosteric binding site, where the APP665–695 sequence binds. These results open the way to attack the allosteric site to prevent APP phosphorylation at Thr654 by ROCK2 without inhibiting the activity of ROCK2 towards its other substrates.

The prevention of preeclampsia is difficult due to the syndromic nature and multiple underlying mechanisms of this severe complication of pregnancy. The current clinical distinction between early- and late-onset disease, although clinically useful, does not reflect the true nature and complexity of the pathologic processes leading to preeclampsia. The current gaps in knowledge on the heterogeneous molecular pathways of this syndrome and the lack of adequate, specific diagnostic methods are major obstacles to early screening and tailored preventive strategies. The development of novel diagnostic tools for detecting the activation of the identified disease pathways would enable early, accurate screening and personalized preventive therapies. We implemented a holistic approach that includes the utilization of different proteomic profiling methods of maternal plasma samples collected from various ethnic populations and the application of systems biology analysis to plasma proteomic, maternal demographic, clinical characteristic, and placental histopathologic data. This approach enabled the identification of four molecular subclasses of preeclampsia in which distinct and shared disease mechanisms are activated. The current review summarizes the results and conclusions from these studies, and the research and clinical implications of our findings.

Protein homodimers have been classified as three-state or two-state dimers depending on whether a folded monomer forms before association, but the details of the folding–binding mechanisms are poorly understood. Kinetic transition networks of conformational states have provided insight into the folding mechanisms of monomeric proteins, but extending such a network to two protein chains is challenging as all the relative positions and orientations of the chains need to be included, greatly increasing the number of degrees of freedom. Here, we present a simplification of the problem by grouping all states of the two chains into two layers: a dissociated and an associated layer. We combined our two-layer approach with the Wako–Saito–Muñoz–Eaton method and used Transition Path Theory to investigate the dimer formation kinetics of eight homodimers. The analysis reveals a remarkable diversity of dimer formation mechanisms. Induced folding, conformational selection, and rigid docking are often simultaneously at work, and their contribution depends on the protein concentration. Pre-folded structural elements are always present at the moment of association, and asymmetric binding mechanisms are common. Our two-layer network approach can be combined with various methods that generate discrete states, yielding new insights into the kinetics and pathways of flexible binding processes.

Miscarriages affect 50-70% of all conceptions and 15-20% of clinically recognized pregnancies. Recurrent pregnancy loss (RPL, >/=2 miscarriages) affects 1-5% of recognized pregnancies. Nevertheless, our knowledge about the etiologies and pathophysiology of RPL is incomplete, and thus, reliable diagnostic/preventive tools are not yet available. Here, we aimed to define the diagnostic value of three placental proteins for RPL: human chorionic gonadotropin free beta-subunit (free-beta-hCG), pregnancy-associated plasma protein-A (PAPP-A), and placental growth factor (PlGF). Blood samples were collected from women with RPL (n = 14) and controls undergoing elective termination of pregnancy (n = 30) at the time of surgery. Maternal serum protein concentrations were measured by BRAHMS KRYPTOR Analyzer. Daily multiple of median (dMoM) values were calculated for gestational age-specific normalization. To obtain classifiers, logistic regression analysis was performed, and ROC curves were calculated. There were differences in changes of maternal serum protein concentrations with advancing healthy gestation. Between 6 and 13 weeks, women with RPL had lower concentrations and dMoMs of free beta-hCG, PAPP-A, and PlGF than controls. PAPP-A dMoM had the best discriminative properties (AUC = 0.880). Between 9 and 13 weeks, discriminative properties of all protein dMoMs were excellent (free beta-hCG: AUC = 0.975; PAPP-A: AUC = 0.998; PlGF: AUC = 0.924). In conclusion, free-beta-hCG and PAPP-A are valuable biomarkers for RPL, especially between 9 and 13 weeks. Their decreased concentrations indicate the deterioration of placental functions, while lower PlGF levels indicate problems with placental angiogenesis after 9 weeks.