A hierarchical, data-driven approach to modeling single-cell populations predicts latent causes of cell-to-cell variability. Loos et al

Published: 25 April 2018| Version 1 | DOI: 10.17632/v4886j8j8n.1
Katharina Möller


We hypothesized that nerve growth factor (NGF)-mediated Erk1/2 phosphorylation is differently modulated in subpopulations of primary sensory neurons when neurons are cultured on poly-D-lysine (PDL) versus collagen type I (Col I). For this, dissociated sensory neurons from dorsal root ganglia of male Sprague Dawley rats were cultured overnight on PDL or Col I and were stimulated acutely with NGF (kinetic: 0, 1, 5, 15, 30, 60, 120 min and 20 ng/ml; dose response: 1h and 0, 0.16, 0.8, 4, 20, 100, 500 ng/ml). NGF-induced signaling activity was monitored by detecting phosphorylated Erk1/2 via immunocytochemistry with phospho-specific antibody labelling (NGF kinetic (file: KM14_KM28KM31KM34_NGFkinetic_pErk_ResultsFinal_Cells) and NGF dose response (file: KM14_KM60KM62KM67KM68_NGFdoseresponse_pErk_ResultsFinal_Cells)). Additionally, co-labelling with antibodies against TrkA (receptor of NGF, file: KM14_KM100KM102KM104KM106_NGFdoseresponse_TrkApERK_ResultsFinal_Cells) and Erk1/2 (file: KM14_KM101KM103KM105KM107_NGFdoseresponse_ERKpERK_ResultsFinal_Cells) should determine differences in these pathway components between PDL/NGF and Col I/NGF treated sensory neurons. Cultures were fully digitalized by the Cellomics ArrayScan XTi HCS microscope and analyzed on a single-cell level by automated image analysis algorithms and R-script processing. NGF kinetic and dose response data show a significant amplitude increase of Erk1/2 phosphorylation in Col I treated cultures. pErk/TrkA and pErk/Erk data present an increase in TrkA and Erk1/2 expression in Col I treated cultures. This experimental single-cell data was used to demonstrate the capacity of our novel hierarchical, data-driven approach to modeling single-cell populations. File description (header): PlateID: Plate IDs corresponding to labbook entries of 3-4 replicate experiments Well: corresponding wells of 96-well plate Stain: Staining of wells with corresponding antibody combinations for spill over compensation of fluorescent channels (Ch1, Ch1Ch2, Ch1Ch3, Ch1Ch2Ch3) Cond: treatment condition (PDL: poly-D-lysine; Col I: collagen type I, SpillOver: for data compensation) Comp: applied compound (Ctrl: solvent, NGF: nerve growth factor, Fsk: forskolin (positive control)) Conc: used concentration (ng/ml) or if not applicable tested wells Time: stimulation time (min) Area: Cell size in µm2 Ch1-Ch4: fluorescent intensities of indicated antibody stainings (ch-UCHL1: chicken polyclonal antibody against UCHL1; rb-pErk: rabbit monoclonal antibody against phospho-Erk1/2; mm-UCHL1: mouse monoclonal antibody against UCHL1; go-TrkA: goat polyclonal antibody against TrkA; mo-Erk1/2: mouse monoclonal antibody against ERK1/2)



Uniklinik Koln


Nerve Growth Factor, Fluorescence, Antibody Labeling, Cellular Signaling