Experimental Biophysics of Complex Systems

Head: Assisant Professor Iztok Urbančič

Code: P1-0060

Duration:January 1, 2019-December 31, 2027

January 1, 2015–December 31, 2018, head: Professor Štrancar Janez, PhD

January 1, 2009-December 31, 2014, head: Professor Serša Igor, PhD

January 1, 2004-December 31, 2008, head: Professor Štrancar Janez, PhD

International collaboration: 

Institute of Biophysics and Nanosystems Research (A)
Helmholtz Munich (DE)
Technische Universität Ilmenau (DE)
Friedrich Schiller University Jena, Leibniz Centre for Photonics in Infection Research, and Jena Center
for Soft Matter, Jena, Germany (DE)
NRCWE – National Research Centre for the Working Environment (DK)
Elettra Sinhrotron Trst (I)
University of Florence (I)
Kyung Hee Univeristy (KR)
Wageningen University (NL)
STAMI – National Institute of Occupational Health (NO)
University of Bergen (NO)
Yale University, School of Medicine, New Haven, CT (ZDA)
Mayo Clinic, Rochester MN (ZDA)
The University of Utah, Salt Lake City UT (ZDA)

Laboratory of Biophysics

MRI Laboratory

Our program group investigates the structure of biological systems and their function in health and
disease by developing and implementing new advanced experimental techniques based around
optical microscopies and magnetic resonance imaging.
In the Laboratory of Biophysics , we study the environmental triggering and early development of
chronic diseases by observing the interactions between (nano)materials, chemicals, and living cells
with advanced optical microscopies, trying to understand how small changes in the functioning of
cell structures trigger and influence the development of diseases. In cooperation with our spin-out
Infinite Biotech, we integrate this knowledge into digital twins that predict the evolution of several
“diseases of civilization”: cancer, neurodegeneration, and chronic lung inflammation. In doing so, we
aim to replace the long and costly animal experiments that are currently used for safety testing of
chemicals and materials, with quick and high-throughput live-cell fluorescence microscopy, image
analysis, automated extraction of the disease mechanism and computer modeling to predict the
material safety and disease development.
To open new possibilities for investigations of living biological systems, we are developing and
implementing advanced optical microscopies, including spectrally-resolved super-resolution STED

and two-photon (2PE) microscopy, multichannel fluorescence lifetime imaging (spFLIM),
fluorescence microspectroscopy (FMS), structured illumination microscopy (SIM), 2.5D projection
imaging, and label-free optical microscopies. To maximize the efficiency of imaging, we are also
developing automated and intelligent autonomous microscopy, and novel fluorescent probes.
In the MRI laboratory , we have developed a method of magnetic resonance imaging with coils with a
nonlinear magnetic field, which are used for spatial signal encoding instead of the usual gradient
coils. The method was first explained theoretically and then experimentally verified.
We introduced a computer simulation procedure based on the chemical bond fluctuation method.
We have successfully used this approach to explain the dissolution processes of tablets based on a
polymer matrix in various media.
We have adapted magnetic resonance imaging to monitor the processes of dendrite growth in
lithium metal batteries. For this purpose, we have also developed a special form of lithium
symmetric cell, which is adapted to magnetic resonance imaging. The growth process was monitored
with this type of cell for various battery charging conditions.
Diffusion tensor magnetic resonance imaging is one of the most advanced methods of medical
imaging diagnostics. We have used this method under magnetic resonance microscopy conditions to
study the diffusion properties of peripheral nerves. We have also studied how these properties
change in various disease states. This research was conducted in collaboration with the Institute of
Anatomy of the Faculty of Medicine of the University of Ljubljana.

In the Laboratory of Biophysics we achieved major breakthroughs in understanding the initiation and
early progression of environmentally-triggered cancer and neurodegeneration. Together with our
spin-out Infinite Biotech, the Yale School of Medicine (USA), and the consortium of the 3 M€ EU
project nanoPASS that we are coordinating, we showed how exposure to diesel exhaust can cause
more aggressive cancers , and developed a cell model that replicates hallmarks of neurodegeneration
after exposure to particulate matter.

Diesel exhaust can disrupt cell division and lead to binuclear cell formation, which is linked to
more aggressive cancers
We have shown that combustion products such as diesel exhaust interfere with cell division, which
can lead to whole-genome duplication and the formation of cells with two or more nuclei, a specific
type of genetic change in cancer cells that can enhance metastasis and lead to more aggressive
tumors. This achievement was the result of monitoring the process of cell division in real time for
over 50 environmental pollutants using an automated live-cell high-throughput approach developed
together with our spin-out Infinite Biotech (Figure 1). Our finding were additionally confirmed by
Prof. Sheltzer’s group from Yale School of Medicine who observed that the mutational signature of
stem cells exposed to polycyclic aromatic hydrocarbons (PAHs) found in diesel exhaust was similar to
mutations found in aggressive cancers with more frequent whole-genome duplication. Our
combined observations were the first observed evidence that environmental pollution drives whole-
genome duplication, and were published in Brown, Nature communications, 2024 .

New in vitro cell model for better understanding of neurodegeneration and preclinical screening of
drug candidates.
We have established an environmentally-induced in vitro cell coculture model of local
neurodegeneration able to exhibit most of the cellular hallmarks of neurodegeneration when
exposed to various environmental particulates and engineered nanoparticles (Figure 2). In contrast,
most currently used rodent models genetically recapitulate protein aggregation seen in the patient
brains, but do not show neuronal degeneration, imposing challenges for testing of drug candidates.
Our model thus offers a promising new platform for mechanistic investigations of causes and
mechanistic progression of Alzheimer’s disease, as well as for preclinical screening of drug
candidates. The results were published in a prominent journal in the field of nanotoxicology research
Nanotoxicology ( Sebastijanović, Nanotoxicology, 2024 ).

Figure 2. Pathologies observed in differentiated neurons (green cytosol) exposed to anatase
TiO 2 nanotubes (red). Left: extracellular plaques containing amyloid-β (blue) and tau protein
(gray); Right: rupture of neuronal connections (yellow arrows).

Development and application of advanced optical microscopy methods: SiM microscopy,
intelligent autonomous microscopy, novel fluorescent probes, and spectral STED microscopy

For the newly established Center of Advanced Optical Microscopies (CNOM), which operates within
the scope of the Center for Electron Microscopy and Microanalysis (CEMM), we successfully
continued building the cutting-edge high-throughput fluorescence microscope, which combines
Structured Illumination microscopy (SIM) (achieving the expected 1.75 resolution improvement over
confocal microscopy) and 2.5D microscopy, and are adding the remaining QPI modality (Figure 3).
The system is fully automated with an easy-to-use user interface, enabling multi-day imaging of a
multitude of samples, including live cells. Our goal is to complete the system by the end of
September 2025 with a hands-on demonstration to the wider community at J. Stefan Institute.

Figure 3. Custom fluorescence microscope that combines super-resolution imaging (SIM),
projection imaging (2.5D) and label-free microscopy (QPI), built in-house in the Center for
Advanced Optical Microscopies.

We are using the algorithms developed within the projects on autonomous microscopy (in
collaboration with the Faculty of Computer and Information Science, University of Ljubljana) for
characterization of sparse nuclear RNA-protein condensates, which appear to be crucial in the
regulation of embryonic development (collaboration with the group of dr. Miha Modic/prof. Jernej
Ule, National Institute of Chemistry, Ljubljana, and Francis Crick Institute, London, UK).
We tested and characterized several novel fluorescent probes and intracellular sensors for advanced
microscopy and microspectroscopy techniques, which were synthesized at the Faculty of Pharmacy
(University of Ljubljana) ( Kokot, Bioorganic chemistry, 2024 ).
With collaborators from University of Oxford (prof. C. Eggeling, now at University of Jena, Germany),
we characterized the unexpected phenomenon of photoblueing, i.e. the process where the
spectrum of a fluorescent dye shifts towards lower wavelengths after being exposed to high laser
powers, used in fluorescence microscopy ( Dasgupta, Nature methods, 2024 ). The detailed
characterization of this phenomenon was enabled by our unique combination of a STED microscope
equipped with a spectral detector.

Dissolution simulation of polymer matrix tablets

Polymer matrix tablets are an important oral drug delivery system today. In this study, magnetic
resonance microscopy (MRM) was used to analyze the water hydration process in xanthan-based
tablets by measuring erosion, swelling, and penetration fronts. These experimental results were
then complemented by numerical simulations of the same process, i.e. the polymer matrix of the
tablet is modeled as an assembly of interacting chains with embedded drug particles, while its
hydration process is mediated by interactions with solvent particles (Figure 4). The swelling
dynamics were modeled using a Monte Carlo-based bond-fluctuation model (BFM), which elegantly
accounts for steric and nearest-neighbor interactions. The study, published in: (Bajd, Polymers,
2024), provides an efficient experimental-theoretical approach to studying polymer matrix swelling
processes.

Fig. 4: A flowchart schematically summarizing the essential steps in (a) magnetic resonance imaging
(MRI) experiments of polymer tablet swelling enabling the determination of erosion, swelling and
penetration fronts (EF, SF, PF) and in (b) explicit-solvent 2D bond fluctuation model (BFM)
simulations of the swelling process of drug-containing polymer matrices.

Magnetic resonance imaging and simulation of dendritic growth in lithium metal batteries

We developed a new simulation model of dendrite growth based on the diffusion-limited
aggregation (DLA) algorithm upgraded with consideration of the electric field, which improved the
model’s ability to simulate realistic dendrite morphologies. The model predictions were also
confirmed by experimental data on dendrite structures measured by sequential 3D MRI (Figure 5).
This approach provides researchers with a practical tool to explore different strategies for inhibiting
dendrite growth, ultimately contributing to the development of safer and more reliable lithium-ion
batteries. The research was published in: (Peklar, Batteries, 2024).

Fig. 5: MRI imaging results; a representative slice showing the development of the dendritic
structure with time (a). A time sequence of volume-rendered images of dendrite binary masks
extracted from the corresponding MR images of the battery (b).

Magnetic resonance imaging using nonlinear magnetic field coils for spatial signal encoding
Previous approaches to MR imaging using nonlinear magnetic field coils for spatial signal encoding
share a complex reconstruction procedure that originates from the interpretation of the signal in the
spatial frequency domain (k-space), which in our opinion is somewhat misleading and suboptimal.
Instead, we propose a reconstruction approach in which the acquired MR signal is generalized and
thus treated as a function of time rather than as a function of k-space. This time-domain signal is
then Fourier transformed into a spectrum, which usually represents a distorted image of the sample.
In a second step, this spectrum is geometrically and intensity corrected according to the known
frequency-to-space (magnetic field-to-space) transformation of the encoding coil. The use of such
coils in MR imaging could have a significant impact on the imaging speed due to the possible design
of such coils with low inductance. In the article (Tušar, Scientific Reports, 2024), we also
experimentally verified the presented theory.
Assessment of dental pulp changes in caries based on MR imaging and histology
In this study, we assessed the correlation between NMR T 2 relaxation times and their variability with
histopathological results of the same patient’s teeth in relation to caries progression. The study

included 52 extracted permanent premolars whose T 2 times were measured byT 2 mapping before
extraction, and caries progression was assessed using the ICDAS scale. We found a positive
correlation between T 2 times and pulp vessel density and a negative correlation between T 2 times and
macrophage density. The results of the study, which was published in: (Tenyi, BMC Oral Health,
2024), confirmed the usefulness of MRI in assessing the actual condition of pulp tissue.
Development of a driven equilibrium spin echo MR imaging sequence
In this study, we aimed to find an efficient solution for fast MR imaging of samples with long NMR
relaxation times T 1 . Namely, most imaging sequences for this purpose have distortions originating
from non-uniform k-space weighting or T 2 * signal weighting. To avoid these problems, we developed
a double spin echo imaging sequence, which achieves almost complete preservation of the
longitudinal magnetization after each signal acquisition, in contrast to the standard spin echo
sequence, where it is zero. This allows us to perform very fast imaging of the aforementioned
samples with low image distortions. This method, published in: (Serša, J. Magn. Reson., 2024), was
theoretically analyzed and verified under the conditions of microscopic MR imaging.
Diffusion tensor imaging of peripheral nerves with magnetic resonance microscopy
We compared three-dimensional (3D) reconstructions of the median and ulnar nerves obtained by
high-resolution tomographic ultrasound (HRUS) and magnetic resonance microscopy (MRM), and
assessed the ability of both methods to visualize intraneural anatomy. We found that 3D HRUS
underestimates nerve and fascicle volume compared to MRM, and also shows fewer fascicles and
approximately half fewer interfascicular connections. The study was published in: (Pušnik, Eur.
Radiol. Exp., 2024).

Training of young experts
In 2024, 2 PhD candidates (Sebastijanović, Dodlek) and 18 MSc students defended their theses.

Selected publications in 2024

1. Peklar, Rok, Mikac, Urška, Serša, Igor. Simulation of dendrite growth with a diffusion-limited
   aggregation model validated by MRI of a lithium symmetric cell during charging. Batteries.
   2024, vol. 10, iss. 10, str. 1-15, ilustr. ISSN 2313-0105. https://www.mdpi.com/2313-
   0105/10/10/352 , DOI: 10.3390/batteries10100352 . [COBISS.SI-ID 210864643 ]
2. Tu[ar, Kaja, Serša, Igor. Use of nonlinear pulsed magnetic fields for spatial encoding in
   magnetic resonance imaging. Scientific reports. 2024, vol. 14, art. 7521, 15 str. ISSN 2045-
3. DOI: 10.1038/s41598-024-58229-x . [COBISS.SI-ID 191262723 ],
4. Bajd, Franci, Mikac, Urška, Mohorič, Aleš, Serša, Igor. The effect of polymer–solvent
   interaction on the swelling of polymer matrix tablets: a magnetic resonance microscopy
   study complemented by bond fluctuation model simulations. Polymers. Mar. 2024, vol. 16,
   iss. 5, [article no.] 601, str. 1-14, ilustr. ISSN 2073-4360. https://www.mdpi.com/2073-
   4360/16/5/601 , Repozitorij Univerze v Ljubljani – RUL , DOI: 10.3390/polym16050601 .
   [COBISS.SI-ID 186456835 ]
5. Brown, L. M. et al. An elevated rate of whole-genome duplications in cancers from Black
   patients. Nat Commun 15, 8218 (2024).
6. Sebastijanović, A. et al. Particulate matter constituents trigger the formation of extracellular
   amyloid β and Tau -containing plaques and neurite shortening in vitro. Nanotoxicology 18,
   335–353 (2024).
7. Dasgupta, A. et al. Effects and avoidance of photoconversion-induced artifacts in confocal
   and STED microscopy. Nature Methods 21, 1171-1174 (2024).