A brief review

This article serves as a brief history and review of EBM—how EBM developed, its strengths and limitations, and the need for constant improvements. Hopefully, this review will have enhanced your understanding of EBM and its importance and stimulated you to apply EBM to your own practice. As more data and therapies become available, and as clinical guidelines continue to evolve based on EBM, we should expect patient outcomes to improve.     

Hospital hypoglycemia: From observation to action

Background: A preponderance of evidence indicates that when treatment of hyperglycemia with insulin is provided for certain hospitalized populations, the attainment of appropriate glycemic targets improves nonglycemic outcomes such as mortality rates, morbidities (eg, wound infection, critical illness polyneuropathy, bacteremia, new renal insufficiency), duration of ventilator dependency, transfusion requirements, and length of hospital stay. Nevertheless, randomized controlled trials (RCTs) of intensive insulin therapy and studies of outcomes before and after implementation of tight glycemic control have consistently recognized an increased incidence of hypoglycemia as a complication associated with the use of lower glycemic targets and higher doses of insulin.Objectives: This commentary compares the quality of the available evidence on the clinical impact of iatrogenic hypoglycemia. We present treatment strategies designed to prevent iatrogenic hypoglycemia in the hospital setting.Methods: The PubMed database and online citations of articles tracked subsequent to publication were searched for articles on the epidemiology, clinical impact, and mechanism of harm of hypoglycemia published since 1986. In addition, we searched the literature for RCTs conducted since 2001 concerning intensive insulin therapy in the hospital critical care setting, including meta-analyses; letters to the editor were excluded. The retrieved studies were scanned and chosen selectively for full-text review based on the study size and design, novelty of findings, and evidence related to the possible clinical impact of hypoglycemia. Reference lists from the retrieved studies were searched for additional studies. Reports were summarized for the purpose of comparing and contrasting the qualitative nature of information about iatrogenic hypoglycemia in the hospital.Results: Eight RCTs of intensive glycemic management, 16 observational studies of hospitalized patients with hypoglycemia (including studies of outcomes before and after implementation of tight glycemic control), and 4 case reports on patients with hypoglycemia were selected for discussion of the incidence of hypoglycemia, significance of hypoglycemia as a marker or cause of poor prognosis, and clinical harm of hypoglycemia. Hypoglycemia was identified in clinical trials as either a category of adverse events or a complication of intensified insulin treatment. For example, a recent meta-analysis found that the incidence of severe hypoglycemia was higher among critically ill patients treated with intensive insulin therapy than among control patients, with a pooled relative risk of 6.0 (95% CI, 4.5–8.0). In the largest multisite RCT on glycemic control among patients in intensive care units (ICUs) conducted to date, deaths were reported for 27.5% (829/3010 patients) in the intensive-treatment group and 24.9% (751/3012 patients) in the conventional-treatment group (odds ratio, 1.14; 95% CI, 1.02–1.28; P = 0.02). In another multisite ICU study, although the intensive and control groups had similar mortality rates, the mortality rate was higher among hypoglycemic participants than among nonhypoglycemic participants (32.2% vs 13.6%, respectively; P < 0.01). Pooled data from 2 singlesite studies in medical and surgical ICUs revealed an increased risk of hypoglycemia in the intensive-treatment group compared with the conventional-treatment group (11.3% [154/1360] and 1.8% [25/1388], respectively; P < 0.001), but the hospital mortality rate was similar for the 2 groups (50.6% [78/154] and 52.0% [13/25], respectively). Specific sequelae of hypoglycemia affecting individual patients were described in the RCTs as well as in the observational studies. New guidelines for glycemic control have recently been issued, but results of the studies using the new targets are not yet available. We propose treatment strategies designed to prevent iatrogenic hypoglycemia in the hospital setting.Conclusions: In response to the growing evidence on the risk of hypoglycemia during intensified glycemic management of hospitalized patients, professional organizations recently revised targets for glycemic control. It is appropriate for institutions to reevaluate hospital protocols for glycemic management with intravenous insulin and, on general wards, to implement standardized order sets for use of subcutaneous insulin to achieve beneficial targets using safe strategies.     

Commentary: When does understanding phenotypic evolution require identification of the underlying genes?

Adaptive evolution is fundamentally a genetic process. Over the past three decades, characterizing the genes underlying adaptive phenotypic change has revealed many important aspects of evolutionary change. At the same time, natural selection is often fundamentally an ecological process that can often be studied without identifying the genes underlying the variation on which it acts. This duality has given rise to disagreement about whether, and under what circumstances, it is necessary to identify specific genes associated with phenotypic change. This issue is of practical concern, especially for researchers who study nonmodel organisms, because of the often enormous cost and labor required to “go for the genes.” We here consider a number of situations and questions commonly addressed by researchers. Our conclusion is that although gene identification can be crucial for answering some questions, there are others for which definitive answers can be obtained without finding underlying genes. It should thus not be assumed that considerations of “empirical completeness” dictate that gene identification is always desirable.     

Targeted disruption of p53 attenuates doxorubicin-induced cardiac toxicity in mice

Use of the chemotherapeutic agent doxorubicin (Dox) is limited by dose-dependent cardiotoxic effects. The molecular mechanism underlying these toxicities are incompletely understood, but previous results have demonstrated that Dox induces p53 expression. Because p53 is an important regulator of the cell birth and death we hypothesized that targeted disruption of the p53 gene would attenuate Dox-induced cardiotoxicity. To test this, female 6–8 wk old C57BL wild-type (WT) or p53 knockout (p53 KO) mice were randomized to either saline or Dox 20 mg/kg via intraperitoneal injection. Animals were serially imaged with high-frequency (14 MHz) two-dimensional echocardiography. Measurements of left ventricle (LV) systolic function as assessed by fractional shortening (FS) demonstrated a decline in WT mice as early as 4 days after Dox injection and by 2 wk demonstrated a reduction of 31± 16% (P < 0.05) from the baseline. In contrast, in p53 KO mice, LV FS was unchanged over the 2 wk period following Dox injection. Apoptosis of cardiac myocytes as measured by the TUNEL and ligase reactions were significantly increased at 24 h after Dox treatment in WT mice but not in p53 KO mice. After Dox injection, levels of myocardial glutathione and Cu/Zn superoxide dismutase were preserved in p53 KO mice, but not in WT animals. These observations suggest that p53 mediated signals are likely to play a significant role in Dox-induced cardiac toxicity and that they may modulate Dox-induced oxidative stress.These two authors equally contributed to this study.     

E6AP ubiquitin ligase mediates ubiquitin‐dependent degradation of peroxiredoxin 1

E6‐associated protein (E6AP) is a cellular ubiquitin protein ligase that mediates ubiquitylation and degradation of tumor suppressor p53 in conjunction with the high‐risk human papillomavirus E6 protein. We previously reported that E6AP targets annexin A1 protein for ubiquitin‐dependent proteasomal degradation. To gain a better understanding of the physiological function of E6AP, we have been seeking to identify novel substrates of E6AP. Here, we identified peroxiredoxin 1 (Prx1) as a novel E6AP‐binding protein using a tandem affinity purification procedure coupled with mass spectrometry. Prx1 is a 25‐kDa member of the Prx family, a ubiquitous family of antioxidant peroxidases that regulate many cellular processes through intracellular oxidative signal transduction pathways. Immunoprecipitation analysis showed that E6AP binds Prx1 in vivo. Pull‐down experiments showed that E6AP binds Prx1 in vitro. Ectopic expression of E6AP enhanced the degradation of Prx1 in vivo. In vivo and in vitro ubiquitylation assays revealed that E6AP promoted polyubiquitylation of Prx1. RNAi‐mediated downregulation of endogenous E6AP increased the level of endogenous Prx1 protein. Taken together, our data suggest that E6AP mediates the ubiquitin‐dependent proteasomal degradation of Prx1. Our findings raise a possibility that E6AP may play a role in regulating Prx1‐dependent intracellular oxidative signal transduction pathways. J. Cell. Biochem. 111: 676–685, 2010. © 2010 Wiley‐Liss, Inc.     

The Evolution of Control and Distribution of Adaptive Mutations in a Metabolic Pathway

In an attempt to understand whether it should be expected that some genes tend to be used disproportionately often by natural selection, we investigated two related phenomena: the evolution of flux control among enzymes in a metabolic pathway and properties of adaptive substitutions in pathway enzymes. These two phenomena are related by the principle that adaptive substitutions should occur more frequently in enzymes with greater flux control. Predicting which enzymes will be preferentially involved in adaptive evolution thus requires an evolutionary theory of flux control. We investigated the evolution of enzyme control in metabolic pathways with two models of enzyme kinetics: metabolic control theory (MCT) and Michaelis–Menten saturation kinetics (SK). Our models generate two main predictions for pathways in which reactions are moderately to highly irreversible: (1) flux control will evolve to be highly unequal among enzymes in a pathway and (2) upstream enzymes evolve a greater control coefficient then those downstream. This results in upstream enzymes fixing the majority of beneficial mutations during adaptive evolution. Once the population has reached high fitness, the trend is reversed, with the majority of neutral/slightly deleterious mutations occurring in downstream enzymes. These patterns are the result of three factors (the first of these is unique to the MCT simulations while the other two seem to be general properties of the metabolic pathways): (1) the majority of randomly selected, starting combinations of enzyme kinetic rates generate pathways that possess greater control for the upstream enzymes compared to downstream enzymes; (2) selection against large pools of intermediate substrates tends to prevent majority control by downstream enzymes; and (3) equivalent mutations in enzyme kinetic rates have the greatest effect on flux for enzymes with high levels of flux control, and these enzymes will accumulate adaptive substitutions, strengthening their control. Prediction 1 is well supported by available data on control coefficients. Data for evaluating prediction 2 are sparse but not inconsistent with this prediction.THEORETICAL research on the process of adaptation has focused primarily on describing the size and number of genetic changes underlying phenotypic change (Fisher 1930; Kimura 1983; Orr 1998, 2002, 2003). By contrast, comparatively little theoretical attention has been given to the question of whether certain genes or types of genes are preferentially involved in the process of adaptation. Yet the current debate over the relative importance of regulatory vs. structural genes in morphological evolution (Hoekstra and Coyne 2007; Stern and Orgogozo 2008) clearly indicates that this question is of interest to evolutionary biologists.One situation in which this question is pertinent is the evolution of characters that are influenced by the concentration of end products of metabolic pathways. Often change in end-product concentration can be achieved by substitutions in any one of several genes in the pathway. One example is the intensity of floral pigmentation. To a first approximation, final pigment concentration, and hence color intensity, can be viewed as being determined by the flux rate down the pigment biosynthetic pathway for a fixed time corresponding to the duration of floral development. More generally, any situation in which flux rate determines phenotype is likely to fall in this category. In such situations, metabolic control theory (MCT) (Kacser and Burns 1973) and similar approaches (Heinrich and Rapoport 1974; Savageau 1976) indicate that changes in flux can be achieved by changing the activity of any enzyme in the pathway. We seek to determine whether and, if so, why enzymes differ in the probability that they contribute to evolutionary change in pathway flux.It has been suggested as a general principle that enzymes with the greatest control over flux will be disproportionately involved in such evolutionary change (Hartl et al. 1985; Eanes 1999; Watt and Dean 2000). This argument is based on the theoretical expectation that the probability of fixation of an advantageous allele is roughly proportional to its selection coefficient (Hedrick 2000). Since mutations equivalent in terms of enzyme kinetic properties will have greater effects on flux, and hence on fitness, in enzymes with greater metabolic control, mutations in those enzymes will be substituted preferentially.While this argument is likely sound, it simply pushes back the question of which genes evolve preferentially to the question of which enzymes are expected to have greatest control over flux. Although we are unaware of any theoretical attempts to model the evolution of flux control, many authors have speculated about where in pathways control is expected to be highest.Kacser and Burns (1973) hypothesized that the magnitudes of flux control exerted by different enzymes may be very similar. This hypothesis was based on the result from MCT that in linear pathways, overall flux control can be shared by all enzymes. Since metabolic pathways often consist of many enzymes, each would be expected to have only a limited potential to influence flux. Subsequent theoretical analysis of this hypothesis demonstrated that a given flux is consistent with many different flux-control distributions, including, at one extreme, equal flux control by all enzymes and, at the other extreme, major control by one or a few enzymes and little control for all others (Mazat et al. 1996). However, the question of which of these possibilities, if any, are likely to be favored by selection has not been addressed.Another hypothesis, the epistatic or synergistic principle, predicts that control will be vested in a single enzyme at any given time, but will shift over time among enzymes (Dykhuizen et al. 1987; Keightley 1989; Bost et al. 2001) According to this hypothesis, starting from equal control among enzymes in the pathway, selection to increase (or decrease) flux will cause the activity of one enzyme to increase (decrease). This change results in a decrease (increase) in flux control for the enzyme that changes and an increase (decrease) in control for the other pathway enzymes, causing control to be unequally shared. While this argument seems plausible, there has been no analysis of whether over time all enzymes have an equal chance of having elevated control.Finally, Eanes'' (1999) review of enzyme polymorphisms found that control is often centered in enzymes at pathway branch points, which constitute the most upstream enzymes of their specific branch. Flowers et al. (2007) also demonstrated that branching enzymes tend to exhibit more adaptive substitutions than downstream enzymes as would be expected under the principle that evolutionary change will be concentrated in enzymes with the largest control coefficients. In addition, evolutionary changes in these enzymes may be favored because they allow organisms to modify flux allocation to alternate functions and track environmental fluctuations. This suggestion is supported by the “branch point effect,” a theoretical demonstration that control coefficients can dramatically shift between enzymes depending on the kinetic rates of the two competing enzymes (LaPorte et al. 1984). However, this study does not address the question of how the distribution of control is likely to evolve, but describes only which distributions of control are mathematically possible. Thus, Eanes (1999, p. 318) concludes his review, stating “all enzymes in [a] contributing pathway may not be equal; determining the rule[s] for these inequalities should be a major goal in studies of enzyme polymorphism.”A control coefficient (CC) indicates the degree to which flux through a pathway is altered by a small change in the activity of an enzyme (see appendix; this is equivalent to the sensitivity coefficient of Kacser and Burns 1973). The “rules” governing the distribution of control coefficients are determined by the biological evolution of metabolic systems. While research demonstrates that there are many possible distributions of control coefficients, none has examined which of these is most likely to evolve. The optimization of metabolic systems has been explored in detail (Heinrich et al. 1991, 1997; Heinrich and Schuster 1998). In these studies, however, the investigators employ as optimization criteria maximizing flux, maximizing transient times, or minimizing metabolic intermediates, criteria whose biological and evolutionary relevance is unclear.In an effort to understand how control is expected to be shared among enzymes and predict which enzymes are most likely to contribute to adaptive genetic changes, we present two models of the evolution of flux control in a simple linear pathway. The first model employs the framework of MCT. Although there have been many challenges to the MCT framework (Savageau 1976, 1992; Cornish-Bowden 1989; Savageau and Sorribas 1989), it should be made clear that our aim is not to construct a precisely parameterized model of a particular biological system, but to use this generalized framework to address a single, critically ignored question: What are the rules governing how control will evolve to be distributed among enzymes? The use of the MCT framework to address questions in evolutionary genetics is firmly established, with investigation focused on the molecular basis of dominance (Kacser and Burns 1981; Keightley 1996a; Phadnis and Fry 2005; but see Bagheri and Wagner 2004), the relationship between metabolic flux and fitness (Dykhuizen et al. 1987; Szathmary 1993), the amount of additive and nonadditive genetic variance in metabolic systems (Keightley 1989), whether this variation can be explained by mutation–selection balance (Clark 1991), and patterns of response of quantitative traits to selection (Keightley 1996b). The second model we examine, saturation kinetics (SK), is based on Michealis–Menten kinetics and enables us to relax one major assumption of MCT: that enzymes are far from saturation.Here we limit our analysis to linear pathways as an initial attempt to examine these issues. We find that for such pathways control coefficients will generally evolve to be unequal; that the magnitude of this inequality depends on the thermodynamic properties, rather than the kinetic properties, of each reaction step; that upstream enzymes tend to evolve higher control coefficients than downstream enzymes; and that upstream enzymes fix advantageous mutations in greater numbers, and those mutations have larger effects than in downstream enzymes.     

Ameliorating role of caffeic acid phenethyl ester (CAPE) against isoniazid-induced oxidative damage in red blood cells

Isoniazid (INH) still remains a first-line drug both for treatment and prophylaxis of tuberculosis, but various organs toxicity frequently develops in patients receiving this drug. We aimed to investigate possible toxic effects of INH on rat red blood cells (RBCs), and to elucidate whether Caffeic acid phenethyl ester (CAPE) prevents a possible toxic effect of INH. Experimental groups were designed as follows: control group, INH group, INH + CAPE group. Compared with the control, the INH caused a significant increase in superoxide dismutase (SOD) activity and malondialdehyde (MDA) levels, and a decrease in glutathione peroxidase (GSH-Px) and catalase (CAT), which are recently used to monitor the development and extent of damage due to oxidative stresses. CAPE administration to INH group ameliorated above changes due to INH.