Jean-Lynce GNANAGO1,2,3,4,5,6, Tony GERGES1,2,3,4,5,6, Laura Chastagnier1,2,4,6,7,8,9,10, Emma Petiot2,4,6,7,8,9,10, Vincent SEMET1,2,4,5,6, Philippe Lombard1,2,3,4,5,6, Christophe Marquette1,2,4,6,7,8,9,10, Michel Cabrera1,2,3,4,5,6, and Simon Auguste Lambert1,2,3,4,5,6
1Université Claude Bernard Lyon 1, VILLEURBANNE, France, 2INSA LYON, VILLEURBANNE, France, 3Ecole Centrale Lyon, Ecully, France, 4CNRS, VILLEURBANNE, France, 5AMPERE UMR 5005, VILLEURBANNE, France, 6Université de Lyon, VILLEURBANNE, France, 73d.FAB, VILLEURBANNE, France, 8CPE Lyon, VILLEURBANNE, France, 9ICBMS, VILLEURBANNE, France, 10UMR 5246, VILLEURBANNE, France
Synopsis
Tissue
engineering for regenerative medecine is a growing field which faces structural
and functional challenges at different scales. Real-time quantitative 3D characterization
of tissues both in vitro and in vivo would help biologists assessing
their methods. Magnetic Resonance Imaging (MRI) offers the possibility to
perform such characterization non-invasively. We propose here a 7T MRI coil
integrated within a perfusion tissue engineering bioreactor to perform tissue
assessments throughout its growth. The MRI bioreactor is built using 3D
printing and plastronics. This works resulted in a successful observation of a
bioprinted tissue with a 75 µm in plane resolution.
Introduction
Tissue
engineering and regenerative medecine have been developing for a few decades
now and the number of applications is growing 1 From cartilage to skin, a wide
range of tissues are currently being studied and constructed. However, along
with the increasing number of applications, comes an increasing number of different
parameters of interest for tissue characterization depending on its
application. With multiscale and multiparametric acquisitions, MRI is best
suited for characterizing various types of key properties in tissue
development. Recent advances 2 highlighted the interest of real-time monitoring of tissues during
their growth. As a non-invasive and non-ionizing modality, MRI also qualifies
as a potential candidate for this task. In this context, the development of
dedicated MRI apparatus specifically designed for this field becomes mandatory.
Here we study the feasibility of integrating an MRI sensor directly into the bioreactor
to monitor in real time cell fate and tissue growth in 3D printed
tissue-engineered constructs.Method
Different
MRI setups have been designed for small tissue samples over the years 3,4. The present case is different as
the tissue is living and growing while being probed, to our knowledge, only one
group 5
tackled this issue. This implies the combined integration of a coil and a
non-magnetic bioreactor within the bore of the scanner. Merging these two
functions can be done by addition with a conventional FR4-based circular coil
placed on top of an operating bioreactor. This method is simple to setup, on
the downside, the loss in coil sensitivity due to the bioreactor being placed
between the sample and the coil can be detrimental. In addition, this loss
would be bioreactor-dependent, leading to poor setup transferability between
applications. On the other side, another solution exists with the integration
of the coil within the bioreactor itself. This integration has been made
possible by the previous progresses in plastronics techniques 6. Once the design of the bioreactor
was conceived in SOLIDWORKS, simulations were conducted using CST MICROWAVE
STUDIO. The purpose was to evaluate the influence of the dielectric environment
of the coil on its sensitivity and field homogeneity compared to a conventional
FR4 coil that would be considered as a reference (figure 2).
A 12mm
reception circular coil was designed and integrated within a perfusion bioreactor
based on a previous design 7. This type of bioreactor is commonly used in tissue engineering 8 and fits the scope of this work as
it contains no active parts. The cell culture medium is flowing through to the
sample via 2 inlets connected to external fluidic circulation pumps. The bioreactor
was 3D printed using high temperature resin. The copper coil was then deposited
on the surface of the bioreactor following the method described in 6. X-ray fluorescence measurements
showed a copper thickness of 45µm and a 4-points method placed the resistivity
around 1.75×10-8 Ω.m. As illustrated in figure 1, the coil and its passive decoupling
circuit were conceived in 3D using via through the wall of the bioreactor. The 300
MHz tuning/matching of the coil was done remotely with a tuning box with two
trimmer capacitors, one in parallel (tuning) and one in series (matching).
Finally, a mechanical support was built to allow the bioreactor to fit inside a
36mm radius birdcage transmission coil.
Bench
characterization of the coil included unloaded/loaded quality factor
measurements. For comparison purposes, bench characterization has also been
conducted on a coil made with copper tapped on an FR4 plate. Additional
characterization of the bioreactor coil has been conducted in imaging
conditions with a 7T Bruker MRI scanner. SNR measurements were performed on an echographic
gel using 3D Flash sequence with following parameters: matrix 128×128×64, a FOV
19.2×19.2×9.6 mm3 i.e.150µm3 isotropic resolution, TR
33.81ms, TE 15ms, 16 averages for a total acquisition time of 74min. As proof
of concept for our bioreactor, a 2D turborare MRI acquisition was performed on
a tissue engineering construct with following parameters: matrix 256×256×64, a
FOV 19.2×19.2mm2 i.e. 75µm2 in-plane resolution, 12
slices of 500 µm thickness, TR 3000ms, TE 15ms, Rare factor 8, 4averages for a
total acquisition time of 4min 48sResults
For long TE
sequences, an important image artifact appeared. Numerical simulations showed
that this is due to the 3D structure of the bioreactor coil leading to a
parasitic loop causing field extinction near its location (figure 4). This
parasitic loop could explain that the SNR was 1.36 times lower than the SNR
obtained with a larger commercial single loop coil available at the imaging
facility (16 mm diameter). An in-plane 75 µm² resolution on the bioprinted
sample was achieved with the bioreactor.Conclusion
The 3D integration
of the MRI coil to the bioreactor needs to be improved to open the way for the first real-time
monitoring of a constructing tissue. A validation step for this integration would
require a B1 mapping using recent methods 9. The integration of a piezoelectric vibrating
membrane is also studied for elastography purposes. In conclusion, this work
demonstrates the feasibility of an integrated coil inside a bioreactor and paves
the way for a democratization of the MRI tool for biologists using
tissue-tailored fully integrated MRI- bioreactors.Acknowledgements
This work
was supported by a grant from the Agence National de la Recherche (Estimate
Project N° ANR-18-CE19-0009-01). The financial support provided by Ingénierie@Lyon,
member of the Carnot Institutes Network (Metafab 3D project) for the
postdoctoral scholarship of Dr T. Gerges is also acknowledged. Moreover, the role of CERMEP - Imagerie du vivant and Radu Bolbos and is acknowledged.References
1. Shafiee,
A. & Atala, A. Tissue Engineering: Toward a New Era of Medicine. Annu.
Rev. Med. 68, 29–40 (2017).
2. Welter,
J. F. & Baskaran, H. Monitoring and real-time control of tissue engineering
systems. in Principles of Tissue Engineering 1459–1467 (Elsevier, 2020).
doi:10.1016/B978-0-12-818422-6.00079-4.
3. Meadowcroft,
M. D. et al. Direct magnetic resonance imaging of histological tissue
samples at 3.0T. Magn. Reson. Med. 57, 835–841 (2007).
4. Sengupta,
S. et al. High resolution anatomical and quantitative MRI of the entire
human occipital lobe ex vivo at 9.4 T. NeuroImage 168, 162–171
(2018).
5. Othman,
S. F., Wartella, K., Khalilzad Sharghi, V. & Xu, H. The e
-Incubator: A Magnetic Resonance Imaging-Compatible Mini Incubator. Tissue
Engineering Part C: Methods 21, 347–355 (2015).
6. Gerges,
T. et al. 3D Plastronics for Smartly Integrated Magnetic Resonance
Imaging Coils. Front. Phys. 8, 240 (2020).
7. Pourchet,
L. LARGE 3D BIOPRINTED TISSUE_ HETEROGENEOUS PERFUSION AND VASCULARIZATION. 13.
8. Gaspar,
D. A., Gomide, V. & Monteiro, F. J. The role of perfusion bioreactors in
bone tissue engineering. Biomatter 2, 167–175 (2012).
9. Sacolick,
L. I., Wiesinger, F., Hancu, I. & Vogel, M. W. B1 mapping by Bloch-Siegert
shift. Magn. Reson. Med. 63, 1315–1322 (2010).