Immunohistochemical Verification of Synapses and the Use of Machine Learning for the Analysis of the Synaptoarchitecture of the CA3 Hippocampus in White Rats During the Post-Ischemic Period

This study examines the methodological aspects of applying machine learning algorithms to explore new opportunities in interpreting inter-neuronal connections. The aim was to demonstrate that the combination of Ilastik and StarDist is effective for the morphometric characterization of giant synaptic terminals in the stratum lucidum of CA3 in the hippocampus of white rats under normal conditions and in the post-ischemic period. Material and methods. Cerebral ischemia in Wistar white rats was modeled by bilateral occlusion of the common carotid arteries (OCCA) for 20 minutes. Animals were studied without intervention (n=6, control) and at 6 hours, 1, 3, 7, 14, and 30 days after OCCA (n=36). Staining with hematoxylin and eosin, Nissl staining with thionine, as well as immunohistochemical reaction for synaptophysin, were used. Numerical density (NDT), sizes, staining intensity, and area of the terminals were determined, and the Ilastik and StarDist plugins were applied on the ImageJ/Fiji platform. Statistical analysis was performed using non-parametric methods in Statistica 8.0. Results. The relative area of the terminals did not differ between the manual method and machine learning. Machine learning provided additional information on numerical density, sizes, and average brightness of the terminals. At 6 hours after OCCA, NDT decreased by 44.3%, but then recovered over 7 days. The average area of the terminals was 16.7% larger at 6 hours and 1 day, but smaller than the control level at 14 days. The brightness of the terminal pixels was inversely proportional to the content of chromogen: it increased at 6 hours and 1 day after OCCA, then returned to control levels. Correlations were observed between the area and brightness of the terminals (R=0.78). Conclusion. The use of the combination of Ilastik and StarDist allowed for accurate assessment of numerical density, sizes, shape, relative area, and staining intensity of synaptic terminals in the hippocampus. Compared to the manual method, the application of machine learning provided significantly more information about the terminals in color immunohistochemical images.

1. Avdeev DB, Akulinin VA, Stepanov AS, Gorbunova AV, Stepanov SS. Plejotropnye fermenty apoptoza i sinapticheskaya plastichnost' gippokampa belyh krys posle okklyuzii obshchih sonnyh arterij. Sibirskij medicinskij zhurnal. 2018;33(3):102–110. (In Russ.). doi: 10.29001/2073-8552-2018-33-3-102-110.

2. Borovikov V. Statistica. Iskusstvo analiza dannyh na komp'yutere. 2-oe izd. SPb: Piter; 2003: 688. (In Russ.).

3. Semchenko VV, Stepanov SS, Ereniev SI. Sinapticheskaya plastichnost' golovnogo mozga (fundamental'nye i prikladnye aspekty). M: Direkt-Media; 2024. 416. (In Russ.).

4. Stepanov SS, Koshman IP, Shoronova AYU, Kalinichev AG, Akulinin VA, Stepanov AS, Avdeev DB, Markelova MV. Strukturnye osnovy izmeneniya mezhnejronnoj kommunikacii nejronov polya CA3 gippokampa belyh krys posle tyazheloj cherepno-mozgovoj travmy. Patologicheskaya fiziologiya i eksperimental'naya terapiya. 2021;65(1):22–34. (In Russ.). doi: 10.25557/0031-2991.2021.01.22-34.

5. Stepanov AS, Akulinin VA, Stepanov SS, Avdeev DB, Gorbunova AV. Kommunikaciya nejronov polya CA3 gippokampa golovnogo mozga belyh krys posle ostroj ishemii. Obshchaya reanimatologiya. 2018;12(5):38–49. (In Russ.). doi: 10.15360/1813-9779-2018-5-38-49.

6. Stepanov SS, Makar'eva LM, Akulinin VA, Korzhuk MS, SHoronova AYU, Avdeev DB, Cuskman IG, Slobodskaya AV. Sopostavlenie immunogistohimicheskogo i ul'trastrukturnogo izucheniya reakcii aksonal'nyh terminalej sensomotornoj kory belyh krys na perevyazku obshchih sonnyh arterij. ZHurnal anatomii i gistopatologii. 2022;11(3):65–74. (In Russ.). doi: 10.18499/2225-7357-2022-11-3-65-74.

7. Berg S, Kutra D, Kroeger T, Straehle Ch. et al. Ilastik: interactive machine learning for (bio)image analysis. Nature Methods. 2019;16(12):1226–1232. doi: 10.1038/s41592-019-0582-9.

8. Calhoun ME, Jucker M, Martin LJ, Thinakaran G, Price DL, Mouton PR. Comparative evaluation of synaptophysin-based methods for quantification of synapses. J Neurocytol. 1996;25(1):821–828. doi: 10.1007/BF02284844.

9. Ciresan Dan, Meier U, Schmidhuber J. Multicolumn deep neural networks for image classification. 2012 IEEE Conference on Computer Vision and Pattern Recognition: journal. 2012:3642–3649. doi: 10.1109/cvpr.2012.6248110.

10. Cortada M, Sauteur L, Lanz M, Levano S, Bodmer D. A deep learning approach to quantify auditory hair cells. Hear Res. 2021;15(409):108317. doi: 10.1016/j.heares.2021.108317.

11. Dailey ME, Buchanan J, Bergles DE, Smith SJ. Mossy fiber growth and synaptogenesis in rat hippocampal slices in vitro. J Neurosci. 1994;14(3Pt1):1060–1078. doi: 10.1523/JNEUROSCI.14-03-01060.

12. Fazeli E, Roy NH, Follain G, Laine RF, von Chamier L, Hänninen PE, Eriksson JE, Tinevez JY, Jacquemet G. Automated cell tracking using StarDist and TrackMate. F1000Res. 2020;28(9):1279. doi: 10.12688/f1000research.27019.1.

13. Grys BT, Lo DS, Sahin N, Kraus OZ, Morris Q, Boone C, Andrews BJ. Machine learning and computer vision approaches for phenotypic profiling. J Cell Biol. 2017;2(216(1)):65–71. doi: 10.1083/jcb.201610026.

14. Herbert AD, Carr AM, Hoffmann E. FindFoci: A Focus Detection Algorithm with Automated Parameter Training That Closely Matches Human Assignments, Reduces Human Inconsistencies and Increases Speed of Analysis. PLoS ONE. 2014;9(12):e114749. doi: 10.1371/journal.pone.0114749.

15. Ippolito DM, Eroglu C. Quantifying Synapses: an Immunocytochemistry-based Assay to Quantify Synapse Number. J Visual Experiments. 2010;16(45):2270. doi: 10.3791/2270.

16. Kreshuk A, Zhang C. Machine Learning: Advanced Image Segmentation Using ilastik. Methods Mol Biol. 2019;2040:449–463. doi: 10.1007/978-1-4939-9686-5_21.

17. Kreshuk A, Straehle CN, Sommer C, Koethe U. et al. Automated Detection and Segmentation of Synaptic Contacts in Nearly Isotropic Serial Electron Microscopy Images. PLOS ONE. 2011;6(10):e24899. doi: 10.1371/journal.pone.0024899.

18. Kurnia KA, Sampurna BP, Audira G, Juniardi S, Vasquez RD, Roldan MJM, Tsao CC, Hsiao CD. Performance Comparison of Five Methods for Tetrahymena Number Counting on the ImageJ Platform: Assessing the Built-in Tool and Machine-Learning-Based Extension. Int J Mol Sci. 2022;26(23(11)):6009. doi: 10.3390/ijms23116009.

19. Kurnia KA, Lin YT, Farhan A, Malhotra N, Luong CT, Hung CH, Roldan MJM, Tsao CC, Cheng TS, Hsiao CD. Deep Learning-Based Automatic Duckweed Counting Using StarDist and Its Application on Measuring Growth Inhibition Potential of Rare Earth Elements as Contaminants of Emerging Concerns. Toxics. 2023;8(11(8)):680. doi: 10.3390/toxics11080680.

20. Laux L, Cutiongco MFA, Gadegaard N, Jensen BS. Interactive machine learning for fast and robust cell profiling. PLoS One. 2020;11(15(9)):e0237972. DOI: 10.1371/journal.pone.0237972

21. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 5-th ed. Amsterdam, Boston: Elsevier Academic Press, 2005.

22. Pelkey KA, McBain CJ. Target-cell-dependent plasticity within the mossy fibre-CA3 circuit reveals compartmentalized regulation of presynaptic function at divergent release sites. J Physiol. 2008;586(6):1495–1502. doi: 10.1113/jphysiol.2007.148635.

23. Sommer C, Straehle C, Koethe U, Hamprecht FA. "Ilastik: Interactive learning and segmentation toolkit". 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro. 2011:230–233. doi: 10.1109/ISBI.2011.5872394.

24. Tinevez JY, Perry N, Schindelin J, Hoopes GM, Reynolds GD, Laplantine E, Bednarek SY, Shorte SL, Eliceiri KW. TrackMate: An open and extensible platform for single-particle tracking. Methods. 2017 Feb 15;115:80-90. doi: 10.1016/j.ymeth.2016.09.016.

25. Urru A, González Ballester MA, Zhang C. 2D + Time Object Tracking Using Fiji and ilastik. Methods Mol Biol. 2019;2040:423–448. doi: 10.1007/978-1-4939-9686-5_20.

26. Weigert M, Schmidt U. Nuclei instance segmentation and classification in histopathology images with StarDist. In: Proceedings of the IEEE International Symposium on Biomedical Imaging Challenges (ISBIC), Kolkata, India, 2022. doi: 10.1109/ISBIC56247.2022.9854534.

27. Zhu J, Zou H, Rosset S, Hastie T. Multi-class AdaBoost. Statistics and Its Interface. 2009;2: 349–360. doi: 10.4310/Sll.2009.v2.n3.a8