Philippe Dehottay

Probing the genome-scale metabolic landscape of Bordetella pertussis, the causative agent of whooping cough

ABSTRACT Whooping cough is a highly contagious respiratory disease caused by Bordetella pertussis. Despite widespread vaccination, its incidence has been rising alarmingly, and yet, the physiology of B. pertussis remains poorly understood. We combined genome-scale metabolic reconstruction, a novel optimization algorithm, and experimental data to probe the full metabolic potential of this pathogen, using B. pertussis strain Tohama I as a reference. Experimental validation showed that B. pertussis secretes a significant proportion of nitrogen as arginine and purine nucleosides, which may contribute to modulation of the host response. We also found that B. pertussis can be unexpectedly versatile, being able to metabolize many compounds while displaying minimal nutrient requirements. It can grow without cysteine, using inorganic sulfur sources, such as thiosulfate, and it can grow on organic acids, such as citrate or lactate, as sole carbon sources, providing in vivo demonstration that its tricarboxylic acid (TCA) cycle is functional. Although the metabolic reconstruction of eight additional strains indicates that the structural genes underlying this metabolic flexibility are widespread, experimental validation suggests a role of strain-specific regulatory mechanisms in shaping metabolic capabilities. Among five alternative strains tested, three strains were shown to grow on substrate combinations requiring a functional TCA cycle, but only one strain could use thiosulfate. Finally, the metabolic model was used to rationally design growth media with >2-fold improvements in pertussis toxin production. This study thus provides novel insights into B. pertussis physiology and highlights the potential, but also the limitations, of models based solely on metabolic gene content. IMPORTANCE The metabolic capabilities of Bordetella pertussis, the causative agent of whooping cough, were investigated from a systems-level perspective. We constructed a comprehensive genome-scale metabolic model for B. pertussis and challenged its predictions experimentally. This systems approach shed light on new potential host-microbe interactions and allowed us to rationally design novel growth media with >2-fold improvements in pertussis toxin production. Most importantly, we also uncovered the potential for metabolic flexibility of B. pertussis (significantly larger range of substrates than previously alleged; novel active pathways allowing growth in minimal, nearly mineral nutrient combinations where only the carbon source must be organic), although our results also highlight the importance of strain-specific regulatory determinants in shaping metabolic capabilities. Deciphering the underlying regulatory mechanisms appears to be crucial for a comprehensive understanding of B. pertussiss lifestyle and the epidemiology of whooping cough. The contribution of metabolic models in this context will require the extension of the genome-scale metabolic model to integrate this regulatory dimension.

Monitoring and control of a bioprocess for malaria vaccine production

Based on genetic manipulations, new strains of S. cerevisae are developed, which can be used for the production of pharmaceuticals. In this study, attention is focused on yeast fed-batch cultures dedicated to the production of a malaria vaccine. The efficient operation of this bioprocess requires on-line monitoring and regulation of the ethanol concentration at a low level (so as to maximize biomass productivity). This paper reports on the development of software sensors for the on-line reconstruction of biomass and ethanol, which are based on simple feedforward neural networks making only use of conventional bioprocess instrumentation (stirrer speed, base addition for pH regulation, etc.). This paper also discusses the design of a robust RST control strategy for regulating the ethanol concentration, which ensures setpoint tracking and asymptotic disturbance rejection. Robustification is achieved through the use of Youla parametrisation and on-line adaptation. This control strategy only requires the a priori knowledge about one yield coefficient and one on-line measurement sensor (i.e. an ethanol probe or the proposed software sensor). The software sensor and controller are tested successfully in real-case experimental runs.

High-yield production of recombinant CRM197, a non-toxic mutant of diphtheria toxin, in the periplasm of Escherichia coli

A high cell density fed-batch process was developed for production of recombinant CRM197, a non-toxic mutant of diphtheria toxin widely used as a carrier in polysaccharide-protein conjugate vaccines. Fully soluble recombinant CRM197 was obtained in high yields and with an authentic N-terminus, by targeting the protein to the periplasm of Escherichia coli using the Signal Recognition Particle (SRP)-dependent signal sequence of FlgI. Response Surface Methodology (RSM) was used to optimize the set-points of key process parameters (pH and feed rate at induction). Optimal production of periplasmic CRM197 was found at a slightly basic pH (7.5). The feed rate during induction was positively correlated with the accumulation of unprocessed cytoplasmic CRM197, consistent with limited capacity of the SRP secretion pathway. Decreasing the feed rate to align the protein synthesis rate with the secretion capacity, resulted in minimal production of cytoplasmic CRM197. Besides, the host background was found critical for production of periplasmic CRM197: B834(DE3) was the highest producer (>3 g/L), while BLR(DE3) produced one third less CRM197, and very low yields (290 mg/L) were obtained with HMS174(DE3). The optimized process is robust and linearly scalable, and represents a 20-fold yield improvement compared to a process based on Corynebacterium diphtheriae.

A versatile, non genetically modified organism (GMO)-based strategy for controlling low-producer mutants in Bordetella pertussis cultures using antigenic modulation.

The uncontrolled presence of non-producer mutants negatively affects bioprocesses. In Bordetella pertussis cultures, avirulent mutants emerge spontaneously and accumulate. We characterized the dynamics of accumulation using high-throughput growth assays and competition experiments between virulent and avirulent (bvg(-) ) isolates. A fitness advantage of bvg(-) cells was identified as the main driver for bvg(-) accumulation under conditions of high virulence factor production. Conversely, under conditions that reduce their expression (antigenic modulation), bvg(-) takeover could be avoided. A control strategy was derived, which consists in applying modulating conditions whenever virulence factor production is not required. It has a wide range of applications, from routine laboratory operations to vaccine manufacturing, where pertussis toxin yields were increased 1.4-fold by performing early pre-culture steps in modulating conditions. Because it only requires subtle modifications of the culture medium and does not involve genetic modifications, this strategy is applicable to any B. pertussis isolate, and should facilitate regulatory acceptance of process changes for vaccine production. Strategies based on the same concept, could be derived for other industrially relevant micro-organisms. This study illustrates how a sound scientific understanding of physiological principles can be turned into a practical application for the bioprocess industry, in alignment with Quality by Design principles.

Experimental study of neural network software sensors in yeast and bacteria fed-batch processes

Nowadays, on-line bioprocess monitoring is still a delicate task due to the lack of on-line measurements of the key components of a culture. In this study the use of artificial neural networks (NNs) as a basis to develop software sensors is investigated. Particularly attention is focused on the use of standard signals, such as those coming from pH or oxygen regulation, to infer information on the evolution of biomass or products of yeast and bacteria fed-batch cultures. The selection of informative signals is achieved through principal component analysis (PCA). Radial basis function (RBF) NNs are then used to estimate the component concentrations of interest. This work is based on extensive experimental studies, considering different cell strains and bioreactor scales. The results of our tests demonstrate the flexibility of NN software sensors in industrial environments.