Neuroglobin protects nerve cells from apoptosis by inhibiting the intrinsic pathway of cell death

Abstract

In the past few years, overwhelming evidence has accrued that a high level of expression of the protein neuroglobin protects neurons in vitro, in animal models, and in humans, against cell death associated with hypoxic and amyloid insult. However, until now, the exact mechanism of neuroglobin's protective action has not been determined. Using cell biology and biochemical approaches we demonstrate that neuroglobin inhibits the intrinsic pathway of apoptosis in vitro and intervenes in activation of pro-caspase 9 by interaction with cytochrome c. Using systems level information of the apoptotic signalling reactions we have developed a quantitative model of neuroglobin inhibition of apoptosis, which simulates neuroglobin blocking of apoptosome formation at a single cell level. Furthermore, this model allows us to explore the effect of neuroglobin in conditions not easily accessible to experimental study. We found that the protection of neurons by neuroglobin is very concentration sensitive. The impact of neuroglobin may arise from both its binding to cytochrome c and its subsequent redox reaction, although the binding alone is sufficient to block pro-caspase 9 activation. These data provides an explanation the action of neuroglobin in the protection of nerve cells from unwanted apoptosis.

Fig. 1

Over-expression of neuroglobin reduces sensitivity towards BH3 mimetic-induced cell death. a qRT-PCR (upper panel, n = 3) and western blot (lower panel) analysis of neuroglobin (Ngb) in stable SH-SY5Y-derived cell lines. b Flow cytometry analysis reveals that neuroglobin over-expression partially protects against cell death induced by 2.5 h exposure to 25 μM HA14-1 (mean ± SD, n = 6). Results represent the average from 2 control (pcDNA3.1) cell lines (#2 and #3) and 3 Ngb-over-expressing cell lines (#2, #3, and #4). c Long-term colony survival of wild-type and neuroglobin-over-expressing cells following 48 h treatment with 12.5 μM HA14-1 reveals nearly complete protection by neuroglobin (mean ± SD, n = 10)

Fig. 2

Lowest energy soft-docked structure of the complex between neuroglobin (left) and cytochrome c (right) calculated using BiGGER [24] and rendered using YASARA (www.yasara.org)

Fig. 3

The activation of apoptosome and caspase-9 depend on neuroglobin concentration. a Neuroglobin = 0, b neuroglobin = 0.01 μM, and c neuroglobin = 0.1 μM. Cytochrome c concentration is uniform at 0.1 μM in all three studies. Time is measured in monte carlo (MC) simulations steps. 1 MC step = 10⁻⁴ s, hence time-scale shown 10⁸ MC steps ~2.8 h. Caspase 9 activation is shown (up to 80% completion) normalized by the maximum. Each cross and line represent apoptosome formation and caspase 9 activation, respectively, at a single cell level

Fig. 4

Probability distribution of caspase 3 calculated from single cell activation data of caspase 3. We show results for two neuroglobin concentrations: neuroglobin = 0 (white bars) and neuroglobin = 0.01 μM (black bars). Cytochrome c concentration is fixed at 0.1 μM. Caspase 3 concentration (x-axis) is normalized by the maximum. Time is measured in monte carlo (MC) simulations steps. 1 MC step = 10⁻⁴ s

Fig. 5

Neuroglobin binding to cytochrome c is sufficient to block activation of caspase 9. a Assay of the apoptosome activation of caspase 9 used the cytosol obtained from cultured Jurkat cells to provide the major components of the apoptosome (dATP, Apaf-1 and pro-caspase 9). Apoptosome activation was initiated by the addition of cytochrome c either in the absence or presence of added neuroglobin, and the subsequent assembly of the apoptosome monitored by following the cleavage of a luminescent peptide substrate for caspase 9. b Caspase 3 activation data without the neuroglobin reduction of cytochrome c to a non-apoptotic form. We used cytochrome c 0.1 μM and neuroglobin 0.1 μM in these simulations. Each line shows the time-course of caspase-3 activation at the single cell level. Caspase 3 activation is normalized by the maximum. Time is measured in monte carlo (MC) simulations steps. 1 MC step = 10⁻⁴ s

Fig. 6

Increased levels of cytochrome c require higher levels of neuroglobin to block activation of caspase 3. Time course of caspase-3 activation is shown for three different values of cytochrome c concentrations: a 0.2 μM, b 0.5 μM, c 1 μM. Neuroglobin concentration is kept fixed at 0.1 μM. Each line shows the time-course of caspase-3 activation at the single cell level. Caspase 3 activation is normalized by the maximum. Time is measured in monte carlo (MC) simulations steps. 1 MC step = 10⁻⁴ s. Note the variation in time-scales in caspase 3 activation as the cytochrome c concentration is increased