About MAtCHLESS

Our Focus

Cells: The Complex Units of Life

Eukaryotic and prokaryotic cells are fundamental units of life. To preserve their normal functions, they maintain an internal chemical and physical equilibrium called homeostasis.

Even in this "normal" state, different cell compartments are characterized by different values of parameters such as temperature, oxygen concentration, and pH – all of which are highly interconnected.

However, this equilibrium can get disrupted by the insurgence of conditions like cancer or upon exposure of the cell to external agents.

To thoroughly understand the normal workings of cells and how crippling illnesses and external phenomena impact cell physiology, one should be able to simultaneously map in real time, at the subcellular level multiple chemical and physical parameters.

Luminescence Sensing

Luminescence sensing is a prime tool in the context of cell studies, because of the following aspects:

The sensors (i.e., luminescent species) have a reduced size compared to cells, and thus minimally interfere with biological processes under investigation.
The remote nature of this sensig approach – guaranteed by the use of light to address the sensors – greatly reduces disruption of the biological system.
Mapping of the interested parameters can be achieved with sub-cellular resolution, rather than obtaining single-point measurements or values averaged over the whole cell.

The development of an effective intracellular luminescence sensing approach requires at the very least:

The design of bright, biocompatible luminescent species whose luminescence is sensitive to changes in the target parameter.
The calibration of the luminescent species as sensors, which entails the selection the most appropriate optical readout feature – often, a so-called luminescence intensity ratio (LIR).

However, there are some roadblocks along the path of realizing a reliable approach for intracellular luminescence sensing.

The Roadblocks towards Reliable Intracellular Sensing

Despite the advantages and long history of intracellular luminescence sensing, there are still doubts regarding the reliability of this family of measurement techniques. Take the case of the temperature of mitochondria: Luminescence thermometry seems to hint at temperatures as high as 50 ºC in these organelles. However, physical calculations grounded in bioogical considerations seem to discount this experimental observation as impossible.

What is preventing luminescence sensing – in many instances – from being considered a reliable source of quantitative information about parameters in cells?

Cross-sensitivity. The main source of loss of accuracy (i.e., the deviation between real and measured value) in luminescence sensing comes from the natural inclination of luminescent species of being sensitive to changes in multiple parameters. In the case discussed above, the MitoTracker® dye used to read temperature [target parameter] appears to be sensitive also to viscosity [interfering parameter]. The complexity of the cellular environment is a curse in this sense, since there is a plethora of possible interfering parameters to be considered.

Variability in sensor properties. Often times, the luminescent species used as sensors suffer from batch-to-batch or even intrabatch variability in size, surface chemistry, and composition. This variability translates to optical hetereogenity, which in turn can lead to faulty readouts.

MAtCHLESS: The Solution

MAtCHLESS addresses these roadblocks through an unconventional, holistic approach that entails an action at two levels:

Upstream approach – i.e., sensor optimization. The sensors are optimized to feature multiple luminescence signals with maximized cross-sensitivity towards 3 selected parameters: Temperature, oxygen concentration, and pH. Sensor-to-sensor variabilities in size and spectral properties are minimized, to ensure identical biological fate and sensor performance. Moreover, they are made biocompatible by design, while featuring flexible surface chemistry that enables organelle-specific targeting.
Downstream approach – i.e., signal analysis optimization. The calibration of the sensors is thoroughly performed considering interfering parameters. The high-dimensionality calibration dataset is analyzed with tailored machine learning algorithms that decouple the sensors response towards changes in temperature, oxygen concentration, and pH, while discarding possible interference by other parameters.

The results of these two actions will be used to transform a simple fluorescence microscope into a multispectral imager, via the implementation of ad hoc filters with highly customizable spectral profiles. The imager will enable simultaneous, intracellular readout of the three target parameters with sub-second temporal resolution.

The Strategy

The Objectives

Overarching objective. To develop a luminescence sensing technology capable of reliably and simultaneously mapping multiple parameters (temperature, oxygen concentration, and pH) with high spatiotemporal resolution in biological microenvironments.

Sub-objective 1

Development of a luminescent particle with complex luminescence signal capable of responding simultaneously to different stimuli.

Sub-objective 2

Development of a particle sorting strategy based on size and spectral properties.

Sub-objective 3

Calibration of multiparametric sensors using machine learning algorithms for signal analysis.

Sub-objective 4

Development of a multispectral microscope for fast multiparametric imaging in cells.

Sub-objective 5

Test of the reliability of the developed approach using eukaryotic and prokaryotic cell lines.

The Workpackages (WPs)

Material development

WP1

WP2

WP3

WP4

Particle preparation

Sorting strategy optimization

Sensing approach development

Machine learning-driven calibration

Multispectral setup development

Technology validation

Tests in biological systems

Feebdack

loops

Our Team.

We bring together experts in chemistry, nanotechnology, photonics, biology, and machine learning to achieve the interdisciplinary goals of MAtCHLESS.

Riccardo Marin

Principal Investigator

Araceli de Aquino

Postdoctoral Fellow

Liyan Ming

Postdoctoral Fellow

Arif Muhammad

PhD Student

Our collaborators at the Universidad Autónoma de Madrid.

Dr. Erving Ximendes

Expert in machine learning for sensing

Prof. Patricia Haro-González

Expert in optical tweezers

Prof. Ricardo Ámils

Expert in extremophile microorganisms