Biochemical and genomic regulation of the trehalose cycle in yeast: review of observations and canonical model analysis
Introduction
Trehalose is a non-reducing disaccharide that has been found in bacteria, eukaryotic microorganisms, plants, insects and invertebrates, but so far not in mammals (Benaroudj et al., 2001). It has been studied extensively in the baker's yeast, Saccharomyces cerevisiae, where it was originally thought to serve as a carbohydrate reservoir like glycogen (Panek and Mattoon, 1977; Lillie and Pringle, 1980). Trehalose is now being recognized as a crucial defense mechanism that stabilizes proteins and biological membranes under a variety of stress conditions, including increased temperature, hydrostatic pressure, desiccation, nutrient starvation, osmotic or oxidative stress, and exposure to toxic chemicals (Crowe et al., 1984; Attfield, 1987; van Laere, 1989; Wiemken, 1990; De Virgilio et al., 1994; Hottiger et al., 1994; Reinders et al., 1997; Hounsa et al., 1998; Singer and Lindquist, 1998a; Iwahashi et al., 2000; Benaroudj et al., 2001). The trehalose system is also important for the control of glucose influx during the cellular response to adverse conditions (Thevelein and Hohmann, 1995; Bonini et al., 2000), and the gene of one of its production enzymes (TPS1) shows strong homology with GGS1, a gene that is associated with a glucose-sensing complex and with transport of glucose into the cell (Thevelein and Hohmann, 1995). Based on its unique properties of stabilizing molecules, its mild sweetness, high solubility, low hygroscopicity and, last but not least, a price that has become affordable through genetic modifications of microorganisms, trehalose has become an important target for biotechnology, where it is produced for food manufacturing, vaccine protection in hot climates, and cosmetic products, such as lip sticks (Schiraldi et al., 2002).
The trehalose pathway consists of only a few metabolites, which form a substrate cycle, and is governed by a surprisingly complex control system that comprises several inhibiting or activating signaling mechanisms. The design and operation of the tightly controlled trehalose cycle are difficult to explain with casual reasoning and cannot even readily be addressed with an experimental approach, because targeted changes in the susceptibility of a given enzymatic step to induction, repression, inhibition or activation are not always easy to implement without causing other changes or decreased viability. As an effective alternative, it is shown here how mathematical modeling can shed light on the regulation, design, and operation of a biochemical pathway, as exemplified with the trehalose cycle. To this end, the trehalose cycle is translated into a canonical model according to the tenets of Biochemical Systems Theory (BST; Savageau (1969a), Savageau (1969b)).
The rational construction of the canonical model requires first a review of key observations on the regulatory features of the trehalose cycle in wild type and mutants. Once the model is symbolically and numerically defined, a first baseline analysis assesses whether the model is reasonable, as judged by its steady-state and dynamical properties. Second, model responses computed for simulated environmental conditions of abundant or low-glucose conditions reveal the degree of model consistency with experimental observations. Third, the model allows evaluation of the specific roles of regulatory signals for the proper functioning of the pathway. The primary tool for this part of the analysis is the method of controlled mathematical comparisons (MCMC), which is an integral part of BST (Savageau, 1985; Irvine, 1991; Alves and Savageau, 2000) and explained in a later section.
Following the analysis of the model under normal conditions, the paper addresses the operational details of the cell's response to heat shock. For this purpose, all enzyme activities are altered in accordance with observed changes in gene expression after heat shock (Stanford Database, 2003). The model analysis of this situation is executed in three steps. First, the heat-shock model is again tested with respect to steady-state and dynamical features. Second, the consequences of each alteration in activity are investigated by comparing the heat-shock model with an alternative model without that particular alteration. Third, this controlled comparison allows us to associate specific changes in enzyme activity with consequent physiological responses.
It seems that the model presented here is the first systematic attempt to integrate genomic, biochemical, and physiological information of the trehalose system into one functioning entity. The results and discussion of this integrative analysis illustrate that, even for a small pathway, the simultaneous, quantitative consideration of experimental observations from all three levels of organization is necessary for gaining a clear picture of its intricate inner workings.
Section snippets
Biological background
Trehalose (α-d-glucopyranosyl α-d-glucopyranoside or α,α-1,1-diglucose) is produced in a multi-step process, whose substrate is glucose (Fig. 1). Glucose is converted into glucose 6-phosphate (G6P) which, together with uridine diphosphate glucose (UDPG), leads to the formation of trehalose 6-phosphate (T6P) and subsequently trehalose. Trehalose can be split into two molecules of glucose, thus closing the trehalose cycle: 2 glucose→G6P+UDPG→T6P→trehalose→2 glucose.
Glucose is taken up from the
Data
All data were obtained from publicly available sources. Biochemical and physiological information was retrieved from the original literature and through curated websites, especially Brenda (2003) and YPD (Incyte, 2003). This information was used to establish the flow and regulatory organization of the trehalose cycle and to estimate parameter values for the numerical implementation of the canonical model.
Much of the kinetic information needed for setting up the canonical model was obtained from
Concepts of model design
The main challenge in any modeling effort is the choice of the best-suited model structure. While biochemical tradition often uses Michaelis–Menten rate laws as the basic description of enzyme-catalysed reactions, these functions become unwieldy in larger systems and raise questions about their validity in vivo (cf. Savageau (1992), Savageau (1995); Torres and Voit, 2002). An alternative is the use of power-law rate laws, which form the foundation of BST (Savageau (1969a), Savageau (1969b)) and
Results and discussion
As a direct consequence of the chosen type of analysis, most results are comparative. Some compare the responses of the canonical model with those observed in the living cell, others compare the outputs from two alternative models, one of which represents the actual pathway structure, while the other one represents a similar pathway that differs from the actual pathway in one distinct feature, such as a regulatory signal. To facilitate recognition of which results refer to which system,
Conclusion
Mathematical modeling can be applied to biological phenomena in different ways. Its foremost goal is often seen as making predictions of responses of the modeled entity to untested situations. The analyses shown here have another focus, namely to derive mechanistic and operational explanations for observed regulatory structures and coordinated responses. Casual arguing in terms of causes and effects is often insufficient, even if a system is as small as the trehalose cycle. Similarly, the
Acknowledgements
The author expresses his sincere gratitude to Dr. Yusuf Hannun and Ms. Kellie Sims for critical comments and valuable suggestions that made the mathematical aspects of the article easier to understand. This work was supported in part by a Quantitative Systems Biotechnology grant (BES-0120288) from the National Science Foundation and a Complex Biological Systems grant from the National Institutes of Health (1 R01 GM63265-01; Y.A. Hannun, PI). Any opinions, findings, and conclusions or
References (135)
- et al.
Carbon and energetic uncoupling are associated with block of division at different stages of the cell cycle in several cdc mutants of Saccharomyces cerevisiae
Exp. Cell Res.
(1995) - et al.
Purification and characterization of neutral trehalase from the yeast ABYS1 mutant
J. Biol. Chem.
(1989) Heat-shock response in a yeast tps1 mutant deficient in trehalose synthesis
FEBS Lett.
(1994)Trehalose accumulation in Saccharomyces cerevisiae during response to agents that induce heat shock response
FEBS Lett.
(1987)- et al.
Composition and functional analysis of the Saccharomyces cerevisiae trehalose synthase complex
J. Biol. Chem.
(1998) - et al.
Trehalose accumulation during cellular stress protects cells and cellular proteins from damage by oxygen radicals
J. Biol. Chem.
(2001) - et al.
Trehalose-6-phosphate, a new regulator of yeast glycolysis that inhibits hexokinases
FEBS Lett.
(1993) Metabolic surprises in Saccharomyces cerevisiae during adaptation to saline conditionsquestions, some answers and a model
FEMS Microbiol. Lett.
(2000)- et al.
Effects of temperature shifts on the metabolism of trehalose in Neurospora crassa wild type and a trehalase-deficient (tre) mutant. Evidence against the participation of periplasmic trehalase in the catabolism of intracellular trehalose
Biochim. Biophys. Acta
(1995) - et al.
The biosynthesis of trehalose-6-phosphate
J. Biol. Chem.
(1958)