António E. N. Ferreira

Metrics for GO based protein semantic similarity: a systematic evaluation

BackgroundSeveral semantic similarity measures have been applied to gene products annotated with Gene Ontology terms, providing a basis for their functional comparison. However, it is still unclear which is the best approach to semantic similarity in this context, since there is no conclusive evaluation of the various measures. Another issue, is whether electronic annotations should or not be used in semantic similarity calculations.ResultsWe conducted a systematic evaluation of GO-based semantic similarity measures using the relationship with sequence similarity as a means to quantify their performance, and assessed the influence of electronic annotations by testing the measures in the presence and absence of these annotations. We verified that the relationship between semantic and sequence similarity is not linear, but can be well approximated by a rescaled Normal cumulative distribution function. Given that the majority of the semantic similarity measures capture an identical behaviour, but differ in resolution, we used the latter as the main criterion of evaluation.ConclusionsThis work has provided a basis for the comparison of several semantic similarity measures, and can aid researchers in choosing the most adequate measure for their work. We have found that the hybrid simGIC was the measure with the best overall performance, followed by Resniks measure using a best-match average combination approach. We have also found that the average and maximum combination approaches are problematic since both are inherently influenced by the number of terms being combined. We suspect that there may be a direct influence of data circularity in the behaviour of the results including electronic annotations, as a result of functional inference from sequence similarity.

The glyoxalase pathway: the first hundred years... and beyond.

The discovery of the enzymatic formation of lactic acid from methylglyoxal dates back to 1913 and was believed to be associated with one enzyme termed ketonaldehydemutase or glyoxalase, the latter designation prevailed. However, in 1951 it was shown that two enzymes were needed and that glutathione was the required catalytic co-factor. The concept of a metabolic pathway defined by two enzymes emerged at this time. Its association to detoxification and anti-glycation defence are its presently accepted roles, since methylglyoxal exerts irreversible effects on protein structure and function, associated with misfolding. This functional defence role has been the rationale behind the possible use of the glyoxalase pathway as a therapeutic target, since its inhibition might lead to an increased methylglyoxal concentration and cellular damage. However, metabolic pathway analysis showed that glyoxalase effects on methylglyoxal concentration are likely to be negligible and several organisms, from mammals to yeast and protozoan parasites, show no phenotype in the absence of one or both glyoxalase enzymes. The aim of the present review is to show the evolution of thought regarding the glyoxalase pathway since its discovery 100 years ago, the current knowledge on the glyoxalase enzymes and their recognized role in the control of glycation processes.

Yeast protein glycation in vivo by methylglyoxal Molecular modification of glycolytic enzymes and heat shock proteins

Protein glycation by methylglyoxal is a nonenzymatic post‐translational modification whereby arginine and lysine side chains form a chemically heterogeneous group of advanced glycation end‐products. Methylglyoxal‐derived advanced glycation end‐products are involved in pathologies such as diabetes and neurodegenerative diseases of the amyloid type. As methylglyoxal is produced nonenzymatically from dihydroxyacetone phosphate and d‐glyceraldehyde 3‐phosphate during glycolysis, its formation occurs in all living cells. Understanding methylglyoxal glycation in model systems will provide important clues regarding glycation prevention in higher organisms in the context of widespread human diseases. Using Saccharomyces cerevisiae cells with different glycation phenotypes and MALDI‐TOF peptide mass fingerprints, we identified enolase 2 as the primary methylglyoxal glycation target in yeast. Two other glycolytic enzymes are also glycated, aldolase and phosphoglycerate mutase. Despite enolases activity loss, in a glycation‐dependent way, glycolytic flux and glycerol production remained unchanged. None of these enzymes has any effect on glycolytic flux, as evaluated by sensitivity analysis, showing that yeast glycolysis is a very robust metabolic pathway. Three heat shock proteins are also glycated, Hsp71/72 and Hsp26. For all glycated proteins, the nature and molecular location of some advanced glycation end‐products were determined by MALDI‐TOF. Yeast cells experienced selective pressure towards efficient use of d‐glucose, with high methylglyoxal formation as a side effect. Glycation is a fact of life for these cells, and some glycolytic enzymes could be deployed to contain methylglyoxal that evades its enzymatic catabolism. Heat shock proteins may be involved in proteolytic processing (Hsp71/72) or protein salvaging (Hsp26).

A quantitative model of the generation of Nε-(carboxymethyl)lysine in the Maillard reaction between collagen and glucose

The Maillard reaction between reducing sugars and amino groups of biomolecules generates complex structures known as AGEs (advanced glycation endproducts). These have been linked to protein modifications found during aging, diabetes and various amyloidoses. To investigate the contribution of alternative routes to the formation of AGEs, we developed a mathematical model that describes the generation of CML [ N(epsilon)-(carboxymethyl)lysine] in the Maillard reaction between glucose and collagen. Parameter values were obtained by fitting published data from kinetic experiments of Amadori compound decomposition and glycoxidation of collagen by glucose. These raw parameter values were subsequently fine-tuned with adjustment factors that were deduced from dynamic experiments taking into account the glucose and phosphate buffer concentrations. The fine-tuned model was used to assess the relative contributions of the reaction between glyoxal and lysine, the Namiki pathway, and Amadori compound degradation to the generation of CML. The model suggests that the glyoxal route dominates, except at low phosphate and high glucose concentrations. The contribution of Amadori oxidation is generally the least significant at low glucose concentrations. Simulations of the inhibition of CML generation by aminoguanidine show that this compound effectively blocks the glyoxal route at low glucose concentrations (5 mM). Model results are compared with literature estimates of the contributions to CML generation by the three pathways. The significance of the dominance of the glyoxal route is discussed in the context of possible natural defensive mechanisms and pharmacological interventions with the goal of inhibiting the Maillard reaction in vivo.

Protein glycation in Saccharomyces cerevisiae Argpyrimidine formation and methylglyoxal catabolism

Methylglyoxal is the most important intracellular glycation agent, formed nonenzymatically from triose phosphates during glycolysis in eukaryotic cells. Methylglyoxal‐derived advanced glycation end‐products are involved in neurodegenerative disorders (Alzheimers, Parkinsons and familial amyloidotic polyneurophathy) and in the clinical complications of diabetes. Research models for investigating protein glycation and its relationship to methylglyoxal metabolism are required to understand this process, its implications in cell biochemistry and their role in human diseases. We investigated methylglyoxal metabolism and protein glycation in Saccharomyces cerevisiae. Using a specific antibody against argpyrimidine, a marker of protein glycation by methylglyoxal, we found that yeast cells growing on d‐glucose (100 mm) present several glycated proteins at the stationary phase of growth. Intracellular methylglyoxal concentration, determined by a specific HPLC based assay, is directly related to argpyrimidine formation. Moreover, exposing nongrowing yeast cells to a higher d‐glucose concentration (250 mm) increases methylglyoxal formation rate and argpyrimidine modified proteins appear within 1 h. A kinetic model of methylglyoxal metabolism in yeast, comprising its nonenzymatic formation and enzymatic catabolism by the glutathione dependent glyoxalase pathway and aldose reductase, was used to probe the role of each system parameter on methylglyoxal steady‐state concentration. Sensitivity analysis of methylglyoxal metabolism and studies with gene deletion mutant yeast strains showed that the glyoxalase pathway and aldose reductase are equally important for preventing protein glycation in Saccharomyces cerevisiae.

Quantitative assessment of the glyoxalase pathway in Leishmania infantum as a therapeutic target by modelling and computer simulation

The glyoxalase pathway of Leishmania infantum was kinetically characterized as a trypanothione‐dependent system. Using time course analysis based on parameter fitting with a genetic algorithm, kinetic parameters were estimated for both enzymes, with trypanothione derived substrates. A Km of 0.253 mm and a V of 0.21 µmol·min−1·mg−1for glyoxalase I, and a Km of 0.098 mm and a V of 0.18 µmol·min−1·mg−1 for glyoxalase II, were obtained. Modelling and computer simulation were used for evaluating the relevance of the glyoxalase pathway as a potential therapeutic target by revealing the importance of critical parameters of this pathway in Leishmania infantum. A sensitivity analysis of the pathway was performed using experimentally validated kinetic models and experimentally determined metabolite concentrations and kinetic parameters. The measurement of metabolites in L. infantum involved the identification and quantification of methylglyoxal and intracellular thiols. Methylglyoxal formation in L. infantum is nonenzymatic. The sensitivity analysis revealed that the most critical parameters for controlling the intracellular concentration of methylglyoxal are its formation rate and the concentration of trypanothione. Glyoxalase I and II activities play only a minor role in maintaining a low intracellular methylglyoxal concentration. The importance of the glyoxalase pathway as a therapeutic target is very small, compared to the much greater effects caused by decreasing trypanothione concentration or increasing methylglyoxal concentration.

Protein glycation in vivo: Functional and structural effects on yeast enolase

Protein glycation is involved in structure and stability changes that impair protein functionality, which is associated with several human diseases, such as diabetes and amyloidotic neuropathies (Alzheimers disease, Parkinsons disease and Andrades syndrome). To understand the relationship of protein glycation with protein dysfunction, unfolding and beta-fibre formation, numerous studies have been carried out in vitro. All of these previous experiments were conducted in non-physiological or pseudo-physiological conditions that bear little to no resemblance to what may happen in a living cell. In vivo, glycation occurs in a crowded and organized environment, where proteins are exposed to a steady-state of glycation agents, namely methylglyoxal, whereas in vitro, a bolus of a suitable glycation agent is added to diluted protein samples. In the present study, yeast was shown to be an ideal model to investigate glycation in vivo since it shows different glycation phenotypes and presents specific protein glycation targets. A comparison between in vivo glycated enolase and purified enolase glycated in vitro revealed marked differences. All effects regarding structure and stability changes were enhanced when the protein was glycated in vitro. The same applies to enzyme activity loss, dimer dissociation and unfolding. However, the major difference lies in the nature and location of specific advanced glycation end-products. In vivo, glycation appears to be a specific process, where the same residues are consistently modified in the same way, whereas in vitro several residues are modified with different advanced glycation end-products.

Mining GO Annotations for Improving Annotation Consistency

Despite the structure and objectivity provided by the Gene Ontology (GO), the annotation of proteins is a complex task that is subject to errors and inconsistencies. Electronically inferred annotations in particular are widely considered unreliable. However, given that manual curation of all GO annotations is unfeasible, it is imperative to improve the quality of electronically inferred annotations. In this work, we analyze the full GO molecular function annotation of UniProtKB proteins, and discuss some of the issues that affect their quality, focusing particularly on the lack of annotation consistency. Based on our analysis, we estimate that 64% of the UniProtKB proteins are incompletely annotated, and that inconsistent annotations affect 83% of the protein functions and at least 23% of the proteins. Additionally, we present and evaluate a data mining algorithm, based on the association rule learning methodology, for identifying implicit relationships between molecular function terms. The goal of this algorithm is to assist GO curators in updating GO and correcting and preventing inconsistent annotations. Our algorithm predicted 501 relationships with an estimated precision of 94%, whereas the basic association rule learning methodology predicted 12,352 relationships with a precision below 9%.

The glyoxalase pathway in protozoan parasites

The glyoxalase system is the main catabolic route for methylglyoxal, a non-enzymatic glycolytic byproduct with toxic and mutagenic effects. This pathway includes two enzymes, glyoxalase I and glyoxalase II, which convert methylglyoxal to d-lactate by using glutathione as a catalytic cofactor. In protozoan parasites the glyoxalase system shows marked deviations from this model. For example, the functional replacement of glutathione by trypanothione (a spermidine-glutathione conjugate) is a characteristic of trypanosomatids. Also interesting are the lack of glyoxalase I and the presence of two glyoxalase II enzymes in Trypanosoma brucei. In Plasmodium falciparum the glyoxalase pathway is glutathione-dependent, and glyoxalase I is an atypical monomeric enzyme with two active sites. Although it is tempting to exploit these differences for their potential therapeutic value, they provide invaluable clues regarding methylglyoxal metabolism and the evolution of protozoan parasites. Glyoxalase enzymes have been characterized in only a few protozoan parasites, namely Plasmodium falciparum and the trypanosomatids Leishmania and Trypanosoma. In this review, we will focus on the key features of the glyoxalase pathway in major human protozoan parasites, with particular emphasis on the characterized systems in Plasmodium falciparum, Trypanosoma brucei, Trypanosoma cruzi, and Leishmania spp. We will also search for genes encoding glyoxalase I and II in Toxoplasma gondii, Entamoeba histolytica, and Giardia lamblia.

Re and Tc Tricarbonyl Complexes: From the Suppression of NO Biosynthesis in Macrophages to in Vivo Targeting of Inducible Nitric Oxide Synthase

The in vivo molecular imaging of nitric oxide synthase (NOS), the enzyme responsible for the catalytic oxidation of l-arginine to citrulline and nitric oxide (NO), by noninvasive modalities could provide valuable insights into NO/NOS-related diseases. Aiming at the design of innovative (⁹⁹m)Tc(I) complexes for targeting inducible NOS (iNOS) in vivo by SPECT imaging, herein we describe a set of novel (⁹⁹m)Tc(CO)₃ complexes (2-5) and the corresponding rhenium surrogates (2a-5a) containing the NOS inhibitor N(ω)-nitro-l-arginine. The latter is linked through its α-NH₂ or α-COOH group and an alkyl spacer of variable length to the metal center. The complexes 2a (propyl spacer) and 3a (hexyl spacer), in which the α-NH₂ group of the inhibitor is involved in the conjugation to the metal center, presented remarkable affinity for purified iNOS, being similar to that of the free nonconjugated inhibitor (K(i) = 3-8 μM) in the case of 3a (K(i) = 6 μM). 2a and 3a are the first examples of organometallic complexes that permeate through RAW 264.7 macrophage cell membranes, interacting specifically with the target enzyme, as confirmed by the suppression of NO biosynthesis in LPS-treated macrophages (2a, ca. 30% inhibition; 3a, ca. 50% inhibition). The (⁹⁹m)Tc(I)-complexes 2 and 3, stable both in vitro and in vivo, also presented the ability to cross cell membranes, as demonstrated by internalization studies in the same cell model. The biodistribution studies in LPS-pretreated mature female C57BL6 mice have shown that 2 presented an overall higher uptake in most tissues of the LPS-treated mice compared to the control group (30 min postinjection). This increase is significant in lung (3.98 ± 0.63 vs to 0.99 ± 0.13%ID/g), which is known to be the organ with the highest iNOS expression after LPS treatment. These results suggest that the higher uptake in that organ may be related to iNOS upregulation.