ORIGINAL RESEARCH article
Front. Endocrinol.
Sec. Thyroid Endocrinology
CU Cilia – An Application for Image Analysis by Machine Learning – Reveals Significance of Cysteine Cathepsin K Activity for Primary Cilia of Human Thyroid Epithelial Cells
Provisionally accepted- 1Constructor University, Bremen, Bremen, Germany
- 2Constructor Tech, Sofia, Bulgaria
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Cysteine cathepsins K and L are crucial for proper thyroid function in mice. Inhibition of cysteine peptidases in rodent thyroid epithelial cells in vitro results in loss of primary cilia, highlighting the significance of proteolysis for cilia maintenance. In this study, we used an in vitro-approach to ask whether primary cilia of the normal human thyroid epithelial cell line Nthy-ori 3-1 are affected by cysteine cathepsin inhibition in numbers and/or structure. To interpret hundreds of primary cilia in dozens of high-resolution laser-scanning microscopic images, an objective image analysis approach is essential. Therefore, we developed CU Cilia, i.e., a method for the detection of primary cilia and for segmentation of nuclei by machine learning-based image analysis algorithms. For validation of CU Cilia, the results of nuclei segmentation and primary cilia detection are compared with the image analysis software Cell Profiler™. While manual editing by an image analysis expert enhances the already very good image segmentation results when using Cell Profiler™, the herein described CU Cilia application achieved well-comparable results. CU Cilia even allows non-expert users to perform image analysis with little to no training. CU Cilia uses machine learning by a U-Net neural network simplified architecture and requires only a few images as training datasets. Both approaches, Cell Profiler™ and the CU Cilia application, revealed that the broad-spectrum cysteine peptidase inhibitor E64d and the specific cysteine cathepsin K inhibitor Odanacatib affect primary cilia structure significantly, while their numbers remained unaffected. The results point to an elongation of primary cilia in the Odanacatib-treated Nthy-ori 3-1 thyroid epithelial cells, supporting the notion that cysteine cathepsin K activity is physiologically important not only in murine but also in human thyroid cells.
Keywords: Cysteine cathepsins, image analysis, machine learning, neural network pre-training, Odanacatib, primary cilia, Protease Inhibitors, Thyroid epithelial cells
Received: 05 Mar 2025; Accepted: 10 Nov 2025.
Copyright: © 2025 Rehders, Alekseitseva, Gissoni, Dogru, Popov, Boiarov and Brix. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Andrei Boiarov, andrei.boiarov@constructor.tech
Klaudia Brix, k.brix@jacobs-university.de
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
