In vivo integral feedback control for robust synthetic biology

Biotechnology companies use single cells (bacteria, yeast, or mammalian) as 'cell factories' to produce molecules of use in many different sectors, such as pharmaceuticals, enzymes, biofuels, cosmetics or fragrances. In some cases this means that compounds that were previously produced from non-renewable sources (petroleum) can be produced from renewable sources. In other cases cell factories produce useful compounds that would be impossible, too difficult, or too expensive to produce in other ways (e.g. using chemistry). To date, innovation for biotechnological processes has focused on maximising output, but now the challenge is to use cell factories more efficiently by reducing the required input of energy and nutrients. Moreover, as we learn more about how to design and control living cells, we can begin to envision new exciting potential uses for these 'living machines', especially in the healthcare sector.

In order to do this, we need to be able to engineer living cells that behave controllably in the face of changing conditions. This is what this project aims to achieve. In electronic, mechanical and chemical engineering, robust control is typically accomplished through the use of 'Integral Feedback Control', which is an effective strategy to guarantee robustness to step-like perturbations and uncertainties. This requires an integrator. In a nutshell, the integrator accumulates information about the system's past behaviour and uses it to adjust and improve its activity as more information becomes available. Integral Feedback Control allows, for example, cruise control systems to maintain a car at constant speed irrespective of the slope of the road or the combined weight of the passengers; or the speed of an escalator to remain constant regardless of the number of people using it. In this project, we will design, model, construct and test a biological integrator to implement 'in-vivo robust control'.

A fully (re-)programmable and controllable cell is one of the core long-term objectives of the blossoming field of synthetic biology. However, no biological integrator currently exists. To fill this gap, we will engineer the first in vivo 'plug-and-play' bio-integrator device that can be customised for different applications. To demonstrate the functionality of our bio-integrator device, we will use it to create engineered cells that can robustly maintain the concentration of a chosen small molecule around a specified value. To accomplish this, the cell will be equipped with both the ability to sense the extracellular concentration of the molecule and to synthesise and secrete the molecule itself. A rigorous control design will allow for the secretion rate to change dynamically so as to counteract step-like perturbations in the extracellular concentration of the molecule. This will establish the necessary theoretical and experimental basis for future extension of this research into in vivo environments.

For example, a biological integrator device would make it possible to engineer microbes that reside symbiotically with or within other organisms, and that are able to sense and self-adjust to changing and uncertain external conditions. We anticipate that this in turn could lead to the emergence of a revolutionary new form of medicine that we are calling 'active in vivo medicine', i.e. cells that are implanted in patients and monitor the concentration of disease-related biomolecules (e.g. insulin), modulating their production of these molecules in response to patient need.

In order to investigate how active in vivo medicine might be implemented in real-world conditions, we have integrated into this project a programme of work on 'Responsible Research and Innovation' designed to incorporate the perspectives of a wide range of interested parties into any future development of active in vivo medicine, including: biomedical researchers, clinicians, patient groups, regulators, pharmaceutical firms, and bioethicists.

The availability of automatic control mechanisms that can ensure robust and optimal operation of controlled systems is one of the key factors behind the tremendous advances in engineering fields like transportation, industrial production, and energy. However, we are currently unable to systematically design biological control systems and this is impeding the development of synthetic biology applications for bioprocess engineering, novel therapeutics, and advanced biomedicine. This project will bring synthetic biology one step closer to real-world applications, by demonstrating the feasibility of using Biological Integral Feedback Control in living cells. We envisage that our Biological Integrator Device (BID) could be employed in a variety of commercial applications and thus have designed it with modularity in mind so that inputs and outputs can be modified for different applications. We therefore anticipate that the results of this project will have impact on a wide range of beneficiaries over extended timescales.

In the short term (1 to 5 years) the key impact will be on academic research communities, not only in synthetic biology but also in medical and pharmacology research, control engineering, bio-design automation, and social studies of science. For synthetic biology, delivering academic impact on these associated fields is part of the critical pathway towards economic and societal impact because it is an emerging field that relies on - and contributes to - knowledge and skills from a wide range of disciplines.

In the medium term (5 to 10 years), this project will impact a whole range of commercial sectors that are important for the UK and global economic performance and in particular the industrial biotechnology sector. Firms in this sector rely on the development of bacteria capable of producing biofuels, bulk commodity chemicals, and high-value fine and speciality chemicals for food, cosmetic, and pharmaceutical products. Most biotechnological processes are designed to maximise output without taking robustness of the operating conditions into account. A better strategy would be to employ Integral Feedback Control in order to enable cells to dynamically and robustly control their output around an optimal level for the task at hand (see Case for Support and letters of support from firms for examples).

In the longer term (10 to 50 years), we believe that this research could open up a revolutionary new field of biomedicine, which we have called 'active in vivo medicine'. Live engineered cells or microbes would be inserted into patients in order to produce robust homeostatic regulation, inside their bodies, of the concentration of specific molecules of high clinical importance (e.g. insulin or uric acid). Beneficiaries would be patients, medical practitioners and health services, as well as the biomedical firms involved in the production of engineered cells and associated medical devices.

This project will also serve as an exemplar of how to implement 'Responsible Research and Innovation' (RRI) into the life cycle of a synthetic biology project, by incorporating a workpackage (WP) based on the social science concept of Anticipatory Governance. This WP will integrate the perspectives of a wide range of academic and non-academic stakeholders in order to explicitly investigate the real world prospects of the envisaged applications for Biological Integral Feedback Control (especially active in vivo medicine). This will clarify which applications are considered to be more or less desirable and/or realistic, and will also generate additional ideas about the directions in which this research could be developed. Moreover, this WP will provide evidence-based guidance on how to enhance the governance of emerging biosciences and biotechnologies, which will benefit public bodies involved in the funding, regulation and promotion of research and innovation (e.g. Research Councils, HSE, DEFRA, TSB).