Comparison of NET quantification methods based on immunofluorescence microscopy: Hand-counting, semi-automated and automated evaluations

Timo Henneck; Christina Krüger; Andreas Nerlich; Melissa Langer; Leonie Fingerhut; Marta C Bonilla; Marita Meurer; Sönke von den Berg; Nicole de Buhr; Katja Branitzki-Heinemann; Maren von Köckritz-Blickwede

doi:10.1016/j.heliyon.2023.e16982

Comparison of NET quantification methods based on immunofluorescence microscopy: Hand-counting, semi-automated and automated evaluations

Heliyon. 2023 Jun 3;9(6):e16982. doi: 10.1016/j.heliyon.2023.e16982. eCollection 2023 Jun.

Authors

Timo Henneck¹, Christina Krüger¹, Andreas Nerlich², Melissa Langer¹, Leonie Fingerhut^{1

3}, Marta C Bonilla^{1

3}, Marita Meurer^{1

3}, Sönke von den Berg⁴, Nicole de Buhr^{1

3}, Katja Branitzki-Heinemann^{1

3}, Maren von Köckritz-Blickwede^{1

3}

Affiliations

¹ Institute of Biochemistry, University of Veterinary Medicine Hannover, Foundation, 30559, Hannover, Germany.
² Institute for Microbiology, University of Veterinary Medicine Hannover, Foundation, 30173, Hannover, Germany.
³ Research Center for Emerging Infections and Zoonoses (RIZ), University of Veterinary Medicine Hannover, Foundation, 30559, Hannover, Germany.
⁴ Institute for Zoology, University of Veterinary Medicine Hannover, Foundation, 30559, Hannover, Germany.

Abstract

Formation of neutrophil extracellular traps was first described in 2004, showing that NETs are composed of decondensed chromatin fibers and nuclear and granule components. Free DNA is often used to quantify NETs, but to differentiate NETosis from necrotic DNA-release, immunofluorescence microscopy with NET-specific markers is required. Although evaluation by hand is time-consuming and difficult to standardize, it is still widespread. Unfortunately, no standardized method and only limited software tools are available for NET evaluation. This study provides an overview of recent techniques in use and aims to compare two published computer-based methods with hand counting. We found that the selected semi-automated quantification method and fully automated quantification via NETQUANT differed significantly from results obtained by hand and exhibited problems in detection of complex NET structures with partially illogical results. In contrast to that, trained persons were able to adapt to varying settings. Future approaches aimed at developing deep-learning algorithms for fast and reproducible quantification of NETs are needed.

Keywords: Cell counting; ImageJ; In vitro; NETQUANT; NETs; Quantification.