Time-Resolved Transcriptomics and Constraint-Based Modeling Identify System-Level Metabolic Features and Overexpression Targets to Increase Spiramycin Production in Streptomyces ambofaciens.

Marco Fondi,Benedetta Mattorre,Adelfia Talà,Luisa Siculella,Fabrizio Damiano,Clarissa Consolandi,Giuseppe E De Benedetto,Daniela Fico,Gianluca De Bellis,Eva Pinatel,Mariangela Testini,Clelia Peano,Pietro Alifano

doi:10.3389/fmicb.2017.00835

Abstract

In this study we have applied an integrated system biology approach to characterize the metabolic landscape of Streptomyces ambofaciens and to identify a list of potential metabolic engineering targets for the overproduction of the secondary metabolites in this microorganism. We focused on an often overlooked growth period (i.e., post-first rapid growth phase) and, by integrating constraint-based metabolic modeling with time resolved RNA-seq data, we depicted the main effects of changes in gene expression on the overall metabolic reprogramming occurring in S. ambofaciens. Moreover, through metabolic modeling, we unraveled a set of candidate overexpression gene targets hypothetically leading to spiramycin overproduction. Model predictions were experimentally validated by genetic manipulation of the recently described ethylmalonyl-CoA metabolic node, providing evidence that spiramycin productivity may be increased by enhancing the carbon flow through this pathway. The goal was achieved by over-expressing the ccr paralog srm4 in an ad hoc engineered plasmid. This work embeds the first metabolic reconstruction of S. ambofaciens and the successful experimental validation of model predictions and demonstrates the validity and the importance of in silico modeling tools for the overproduction of molecules with a biotechnological interest. Finally, the proposed metabolic reconstruction, which includes manually refined pathways for several secondary metabolites with antimicrobial activity, represents a solid platform for the future exploitation of S. ambofaciens biotechnological potential.

Highlights

Constraint-based modeling is a widely adopted technique to study microbial metabolic features at the system level
Mycelial growth and spiramycin production were measured during S. ambofaciens American Type Culture Collection (ATCC) 23877 growth in shake flask batch cultivation using complex yeast starch (YS) medium (Figure 1 and Supplementary Data Sheet 2)
Among them we found 30 upregulated genes coding for biosynthetic enzymes for secondary metabolites, most of which involved in butyrolactone, stambomycin, and alpomycin/kinamycin production; while among the downregulated genes 37 participate in spiramycin production

Summary

Introduction

Constraint-based modeling is a widely adopted technique to study microbial metabolic features at the system level. It has proven useful for addressing fundamental issues about the metabolic landscape of a variety of microorganisms, including the elucidation of non-trivial metabolic engineering strategies (King et al, 2015; Zhang and Hua, 2016). Computational methods (e.g., Flux Balance Analysis, FBA) can be adopted to infer the flux distributions within the cell (Orth et al, 2010), and, even more importantly, to study the metabolic reprogramming following environmental perturbations and/or changes in any of the possible cellular information layers (e.g., gene expression). A relevant (recent) achievement in the field of constraint-based modeling is the possibility to integrate gene expression data into the metabolic framework. Despite the fact that a recent work showed that no method performs universally well, the integration of transcriptomics data and constraint-based metabolic modeling might still provide cues to guide the determination of the correct phenotype among the space of solutions (Machado and Herrgard, 2014)

Methods

Results

Conclusion