Back to search

IKTPLUSS-IKT og digital innovasjon

Internet of Bio-NanoThings for Prediction and Prevention of Infectious Diseases

Alternative title: Bionano-tingenes internett for prediksjon og prevensjon av smittsomme infeksjoner

Awarded: NOK 16.0 mill.

The focus of the research in 2023 in this project was investigating the basis for developing a Bio-Nano sensor for early detection of bacterial infections. This sensor development will include genetically modified bacteria that can sense a specific analyte, such as a quorum sensing (QS) signal. When the bacteria detect the analyte, a measurable light signal is produced through bioluminescence or fluorescence. To develop reliable quantitative models for predicting the response of the bacterial sensor to the signal, it is crucial to closely examine the complex biochemical processes involved. These processes encompass the uptake of molecules into the bacterial cells and the expression of genes (e.g., LuxABCDE, firely luciferase) responsible for bioluminescence. Through detailed investigations, the project aims to derive accurate models that describe and predict how the bacterial sensor responds to the signal/QS molecules. Bacterial reporter strains and quorum sensing (QS) signals play a crucial role in understanding microbial communication and its impact on various biological processes. In 2023 most resources within the theoretical modeling work package of the CLIPEUS project went into further investigation of bacterial sensors in general. This is mostly in line with the research done in 2022, however, the focus was mostly on the continued analytical examination and mathematical modeling of bacteria-based sensors, in contrast to the anticipated shift of attention towards microfluidic set-ups and integration of bacteria into a practical sensor. To this end various approaches have been pursued: in addition to custom mathematical models in form of ordinary differential equations (ODEs) derived from Mass-Action principles describing the molecular dynamics in a bacterial sensor, alternative modeling techniques were also applied. This was done by employing software solutions like VCell, COPASI or MCell which are well-maintained and supported tools in the fields of systems- and computational- biology. Using these programs, it is possible to build models of bio-chemical systems quickly and efficiently including the investigated bacteria-based sensor, and subsequently simulate the systems both deterministically and stochastically. The use of this approach gives the developed models credibility. In a further attempt to extract useful information from an analytic modeling approach, ideas from Biochemical-Systems-Theory (BST) were used to construct representations of the sensor model which enables facilitated steady-state treatment and in addition allows for analysis of the system dynamics in relation to their sensitivity versus model parameters. This information can be used to identify possibilities for optimizing and engineering bacterial sensors with focus on crucial sensor aspects like sensitivity and responsivity. Following the initial project proposal, further resources were spent on the integration of data-based modeling methods. To this end, Machine-Learning (ML) models were implemented that approximate the functional relationship between analyte concentration and sensor response that do not require domain knowledge, i.e., insight about the precise molecular mechanisms of a bacterial sensor. It is planned to integrate the data-aided modeling approaches into the analytic efforts in form of Physics-Informed-Neural-Networks (PINNs). In 2023 most resources within the experimental modeling work package of the CLIPEUS project went to further advance the project work towards the development of the bacterial-based sensor, genetically engineered reporter strains of Streptococcus mutans and the major human pathogen Streptococcus pneumoniae were employed in time series gene-expression luminescence assays. Wet-lab experimental data provided insight into signal responses upon addition of three important QS molecules: autoinducer 2 (AI-2), competence stimulating peptide (CSP) and sigX inducing peptide (XIP). AI-2 is the only signaling molecule recognized by Gram-positive and Gram-negative bacteria and allows different species to communicate and coordinate their behavior. In S. pneumoniae, QS regulated by AI-2 and the species-specific CSP are involved in formation of biofilms and spread from the nasopharynx to the lungs. Thus, studying these signals both together and individually allows for a comprehensive exploration of bacterial communication and coordination, and provide a deeper understanding of the intricate signaling networks that govern bacterial behavior and phenotypic outcomes. Such knowledge is of vital importance in the development of a bacteria-based sensor and will be extensively applied in the mathematical models.

Our ability to miniaturize sensors and electronics is unprecedented, and our understanding of biological systems enables fine-grained manipulation of cellular behavior down to the molecular level. This project will leverage the PIs’ unique combination of expertise at the crossroads of biology, bio-sensor interface design, and wireless communications, to address the challenges for human health applications such as prediction and prevention of infections. CLIPEUS (Shield in Latin) aims at creating a communications network among man-made implantable devices and the natural cells inside the body under the novel communication paradigm, called Internet of Bio-NanoThings, where sensing, actuation, and computing processes are tightly coupled to pervasively, perpetually, and precisely sense, process, control, and exchange health information of the body in real time to defend it against infections. CLIPEUS focuses on the detection of infections inside the body based on the detection of the molecular communication among the infectious bacteria by man-made bio-nanothings, called GLADIO (Sword in Latin), consisting of bionanosensors for detection, low power electronics for processing and antenna for near field communication to transfer the data through the tissues to outside of the body. Due to the power limitations of GLADIO, the inherent noise of biological processes, and the non-deterministic response of patients to infections, the collected data will be sparse and noisy. Novel machine learning techniques will be developed to interpret this data for future personalized medicine applications. The system will be extensively tested and ameliorated by phantom experiments as a first step before in vivo experiments. The project will recruit and train 2 PhD students and 2 Postdocs and has drawn up a comprehensive, multifaceted plan to disseminate, communicate and exploit the project results.

Funding scheme:

IKTPLUSS-IKT og digital innovasjon