Page 172 - IJB-8-2

SHORT COMMUNICATIONS

Deep Learning-Assisted Nephrotoxicity Testing with

Bioprinted Renal Spheroids

Kevin Tröndle *, Guilherme Miotto , Ludovica Rizzo , Roman Pichler , Fritz Koch , Peter Koltay ,
3
4
2
1
1
1
Roland Zengerle , Soeren S. Lienkamp , Sabrina Kartmann , Stefan Zimmermann 1
1,2
3
1,2
1 University of Freiburg, IMTEK - Department of Microsystems Engineering, Freiburg, 79110, Germany
2 Hahn-Schickard, Freiburg, 79110, Germany
3 Institute of Anatomy, University of Zurich, Zurich, Switzerland
4 Renal Division, Department of Medicine, Faculty of Medicine and Medical Center - University of Freiburg, Freiburg, Germany
Abstract: We used arrays of bioprinted renal epithelial cell spheroids for toxicity testing with cisplatin. The concentration-
dependent cell death rate was determined using a lactate dehydrogenase assay. Bioprinted spheroids showed enhanced
sensitivity to the treatment in comparison to monolayers of the same cell type. The measured dose-response curves revealed an
inhibitory concentration of the spheroids of IC = 9 ± 3 µM in contrast to the monolayers with IC = 17 ± 2 µM. Fluorescent
50
50
labeling of a nephrotoxicity biomarker, kidney injury molecule 1 indicated an accumulation of the molecule in the central
lumen of the spheroids. Finally, we tested an approach for an automatic readout of toxicity based on microscopic images
with deep learning. Therefore, we created a dataset comprising images of single spheroids, with corresponding labels of the
determined cell death rates for training. The algorithm was able to distinguish between three classes of no, mild, and severe
treatment effects with a balanced accuracy of 78.7%.
Keywords: Bioprinting; Spheroids; Kidney; Nephrotoxicity; Deep learning

*Correspondence to: Kevin Tröndle, University of Freiburg, IMTEK - Department of Microsystems Engineering, Freiburg, 79110, Germany;
kevintroendle@gmail.com

Received: November 2, 2021; Accepted: January 19, 2022; Published Online: January 19, 2022
Citation: Tröndle K, Miotto G, Rizzo L, et al., 2022, Deep Learning-Assisted Nephrotoxicity Testing with Bioprinted Renal Spheroids. Int J
Bioprint, 8(2):528. http://doi.org/10.18063/ijb.v8i2.528

1. Introduction since they are derived by reprogramming fibroblasts, an
accessible cell source. Besides the cell type, the structure
The development of novel three-dimensional (3D) cell of the cell models was found to significantly influence
culture models is motivated by their better accuracy in their functionality [1-3] . The simplest cell models are two-
predicting the physiological response of a target organ
in vitro . This would be beneficial for a variety of dimensional (2D) monolayers. On top of this, the structural
[1]
applications, including preclinical drug testing for toxicity complexity could be increased by embedding cells in
or personalized treatment optimizations. In this context, artificial 3D scaffolds to mimic the natural extracellular
the kidney plays a crucial role. Many substances show matrix (ECM). In various tissue engineering studies,
nephrotoxic side effects in late stage clinical studies, which the embedded cells showed unique mechanisms of self-
were not covered in preclinical screenings . To model the assembly and formed complex 3D structures over time,
[2]
[3]
kidney, specific cell types were isolated or reprogrammed including hollow spheroids [4,5] and tubules , both of which
to provide basic characteristics of the cells found in the recapitulated nephron tubule organization and functionality.
functional units of the kidney, that is, the nephric tubules. Direct comparisons with 2D monolayers revealed
These include the frequently used renal proximal tubule an increased sensitivity to treatment with the known
epithelial cells (RPTECs ), or the more sophisticated nephron-toxicant cisplatin, which is a common reference
[3]
induced renal epithelial cells (iRECs ). The latter could substance . Bioprinting was established as an enabling
[3]
[4]
be used prospectively to establish personalized testing, technology for the biofabrication of 3D cell culture models
© 2022 Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution License, permitting distribution and
reproduction in any medium, provided the original work is properly cited.
164