Fig 1.
Quantification of the absolute Reelin amount in supernatants used for stimulation.
Standard curve of recombinant Reelin protein and several dilutions of Reelin-conditioned supernatant, measured by Western blotting. Lanes ‘undiluted’ and ‘1:10’ were omitted from calculation. Measurements comprise three biological replicates.
Fig 2.
Time course of Reelin stimulation in neurons.
Activation and degradation of core components of lipoprotein receptor-dependent Reelin signaling were measured by use of quantitative western blotting. Representative blots of tyrosine-phosphorylation of the neuronal adaptor protein Dab1, total Dab1, total VLDLR and total ApoER2, activation of Src family kinases as well as activation of the PI3 kinase targeting molecule Akt/PKB in primary cortical neurons prepared from E15.5 mice are shown for (A) 0–30 min and (B) 0–240 min of Reelin stimulation.
Fig 3.
Scheme of the monomer (A) and complex (B) signaling models.
In the monomer model variant (A), Dab1 is activated by tyrosine phosphorylation after binding of Reelin to the receptor. The initial phosphorylation of Dab1 initiates phosphorylation of SFKs, which results in a positive feedback. Subsequently, p-Dab1 induces activation of downstream targets, such as Akt. Signaling is negatively regulated by degradation of phosphorylated Dab1, which is further increased through the Reelin-dependent activation of SFKs. In the complex model (B), clusters of the lipoprotein receptors are formed. These bind the adaptor protein Dab1, which is phosphorylated by SFKs. As a feed-forward loop, SFKs activated in a p-Dab1/SFK complex can trans-phosphorylate other Dab1 proteins bound to the receptor complex. The pathway is regulated by ubiquitination and degradation of phosphorylated Dab1.
Fig 4.
Model fits of the monomer and complex models.
Data points are shown as black dots with corresponding uncertainty, the model trajectories as red dashed line (monomer model) or black solid line (complex model). (A) Data and model dynamics after stimulation with a saturating concentration of Reelin. (B) Data and model trajectories after pretreatment with a pan-SFK inhibitor. (C) Reelin stimulation after SFK inhibition for 30 minutes. (D) Data and model response for a dose-response measurement of Reelin at t = 10 min. Reelin concentration is shown in log-space on the x-axis in arbitrary units, with 0 corresponding to the saturating concentration. (E) Dynamics of the model components for phosphorylated Dab1 after trans-phosphorylation (pDab1_trans) or after direct phosphorylation succeeding SFK binding (pDab1_SFK_bound).–[au]: arbitrary units
Fig 5.
Overview of model reduction steps.
(A) Parameter profile likelihood of the Akt deactivation. (B) Dependency of Akt activation and deactivation leading to a coupling with phosphorylated Dab1. (C) Parameter profile of Dab1 trans-phosphorylation that does not exceed the 95% threshold for high values. (D) Remaining model parameters are unchanged in the upper limit. (E) Parameter profile of SFK degradation, which is non-identifiable towards zero. (F) Remaining model parameters re-optimized in turn.
Fig 6.
Time-course of tyrosine-phosphorylated Dab1 after Reelin stimulation in neurons from WT, Vldlr -/- and ApoER2 -/- knockout mice, measured on a single Western blot.
(A) Data is shown as dots with their respective error bars from N = 4 biological replicates. Model trajectories as lines from the complex model that is calibrated on all available data. Representative Westernblots for all conditions are shown on the right. Detailed information about the knockout mice and antibodies are given in Section Materials and Methods. (B) Parameter profile of the amount of ApoER2 within the sum of all receptors, which is compatible with 100% denoting that no Vldl receptors need to be involved in the signaling.
Fig 7.
(A) Prediction bands for an EC50 concentration of Reelin for both monomer (red dashed lines) and complex (black solid lines) models. Data is shown as open circles with errors. The shaded areas correspond to the standard deviation of the model predictions. (B) Model trajectories for Reelin stimulus after SFK inhibition, if SFK-dependent trans-phosphorylation between Dab1 proteins in the cluster is prohibited.