The Timing and Extent of Motor Neuron Vulnerability in ALS Correlates with Accumulation of Misfolded SOD1 Protein in the Cortex and in the Spinal Cord

Abstract

Understanding the cellular and molecular basis of selective vulnerability has been challenging, especially for motor neuron diseases. Developing drugs that improve the health of neurons that display selective vulnerability relies on in vivo cell-based models and quantitative readout measures that translate to patient outcome. We initially developed and characterized UCHL1-eGFP mice, in which motor neurons are labeled with eGFP that is stable and long-lasting. By crossing UCHL1-eGFP to amyotrophic lateral sclerosis (ALS) disease models, we generated ALS mouse models with fluorescently labeled motor neurons. Their examination over time began to reveal the cellular basis of selective vulnerability even within the related motor neuron pools. Accumulation of misfolded SOD1 protein both in the corticospinal and spinal motor neurons over time correlated with the timing and extent of degeneration. This further proved simultaneous degeneration of both upper and lower motor neurons, and the requirement to consider both upper and lower motor neuron populations in drug discovery efforts. Demonstration of the direct correlation between misfolded SOD1 accumulation and motor neuron degeneration in both cortex and spinal cord is important for building cell-based assays in vivo. Our report sets the stage for shifting focus from mice to diseased neurons for drug discovery efforts, especially for motor neuron diseases.

Keywords: ALS; corticospinal motor neuron; misfolded SOD1; motor neurons; selective vulnerability.

Figure 1

Generation of hSOD1^G93A-UeGFP amyotrophic lateral sclerosis (ALS) reporter mouse model and detection of misfolded SOD1 protein. (a) hSOD1^G93A-UeGFP mice were generated by breeding hSOD1^G93A male mice carrying high copy number of human SOD1 protein with a point mutation at position 93 with female UCHL1-eGFP reporter mice that label corticospinal motor neurons (CSMN) with eGFP expression; (b,c) B8H10 antibody detects misfolded human SOD1 protein in the primary motor cortex of hSOD1^G93A-UeGFP (c) but not WT-UeGFP (b) mice. Misfolded SOD1 signal is the brightest in layer 5 where GFP+ CSMN are located. Scale bar, 250 μm.

Figure 2

Misfolded SOD1 protein in the primary motor cortex. (a) B8H10 antibody does not detect any misfolded SOD1 protein in the primary motor cortex of WT-UeGFP control mice; (b–e) misfolded SOD1 protein can be detected in the primary motor cortex of hSOD1^G93A-UeGFP mice by the B8H10 antibody at P30 (b), P60 (c), P90 (d), and P140 (e). Boxed areas enlarged in the right panels. Scale bar, 250 μm (left, low mag) and 50 μm (right, high mag).

Figure 3

Misfolded SOD1 protein in the ventral horn of the lumbar spinal cord. (a) No B8H10 signal is detected in the spinal cord of WT-UeGFP control mice; (b) misfolded SOD1 protein can be detected in the lumbar spinal cord of hSOD1^G93A-UeGFP mice by the B8H10 antibody in ChAT+ SMN. Arrows point to eGFP+ ChAT+ SMN without a misfolded SOD1 signal. Boxed areas enlarged in the panels below. Scale bar, 100 μm (top, low mag) and 50 μm (bottom, high mag).

Figure 4

Misfolded SOD1 protein in the ventral horn of the lumbar spinal cord. (a) There is no misfolded SOD1 protein in the spinal cord of WT-UeGFP control mice; (b–e) misfolded SOD1 protein are detected in the lumbar spinal cord of hSOD1^G93A-UeGFP mice by the B8H10 antibody and degeneration resistant eGFP+ SMN do not have misfolded SOD1 at P30 (b), P60 (c), P90 (d), and P140 (e). Boxed areas enlarged in the right panels. Scale bar, 100 μm (left, low mag) and 50 μm (right, high mag).