Sequential Metabolic Phases as a Means to Optimize Cellular Output in a Constant Environment

Temporal changes of gene expression are a well-known regulatory feature of all cells, which is commonly perceived as a strategy to adapt the proteome to varying external conditions. However, temporal (rhythmic and non-rhythmic) changes of gene expression are also observed under virtually constant external conditions. Here we hypothesize that such changes are a means to render the synthesis of the metabolic output more efficient than under conditions of constant gene activities. In order to substantiate this hypothesis, we used a flux-balance model of the cellular metabolism. The total time span spent on the production of a given set of target metabolites was split into a series of shorter time intervals (metabolic phases) during which only selected groups of metabolic genes are active. The related flux distributions were calculated under the constraint that genes can be either active or inactive whereby the amount of protein related to an active gene is only controlled by the number of active genes: the lower the number of active genes the more protein can be allocated to the enzymes carrying non-zero fluxes. This concept of a predominantly protein-limited efficiency of gene expression clearly differs from other concepts resting on the assumption of an optimal gene regulation capable of allocating to all enzymes and transporters just that fraction of protein necessary to prevent rate limitation. Applying this concept to a simplified metabolic network of the central carbon metabolism with glucose or lactate as alternative substrates, we demonstrate that switching between optimally chosen stationary flux modes comprising different sets of active genes allows producing a demanded amount of target metabolites in a significantly shorter time than by a single optimal flux mode at fixed gene activities. Our model-based findings suggest that temporal expression of metabolic genes can be advantageous even under conditions of constant external substrate supply.

基因表达的时序变化是所有细胞共有的经典调控特征，通常被视为使蛋白质组适配外界环境波动的调控策略。然而，即便在外界环境近乎恒定的条件下，同样可观测到基因表达的时序变化——包括节律性与非节律性两类。本研究提出如下假说：此类时序变化可提升代谢产物的合成效率，优于基因活性恒定状态下的合成表现。为验证该假说，本研究采用了细胞代谢的通量平衡模型（flux-balance model）。将合成特定靶标代谢物（target metabolites）所需的总时长划分为一系列更短的时间区间（即代谢时相），每个区间内仅特定分组的代谢基因处于激活状态。相关通量分布的计算遵循如下约束条件：基因仅存在激活与失活两种状态，且激活基因对应的蛋白质总量仅由激活基因的数量决定——激活基因数目越少，可分配给携带非零通量的酶的蛋白质份额就越多。这种以蛋白质限制为主导的基因表达效率概念，与其他基于“最优基因调控可精准为所有酶与转运蛋白分配所需蛋白质比例，避免速率限制”假设的调控模型存在显著差异。将该概念应用于以葡萄糖或乳酸为替代底物的简化中心碳代谢网络（central carbon metabolism）后，本研究证实：相较于固定基因活性下的单一最优稳态通量模式，通过在包含不同激活基因集合的最优稳态通量模式间切换，可在显著更短的时间内合成所需靶标代谢物。本研究基于模型的结果表明，即便在外界底物供给恒定的条件下，代谢基因的时序表达依然具备优势。