4-Phenylbutyric acid mitigates ER stress-induced neurodegeneration in the spinal cords of a GM2 gangliosidosis mouse model

Abstract

Sandhoff disease (SD), a fatal and rare lysosomal storage disorder (LSD), is caused by a deficiency of the enzyme β-hexosaminidase B and leads to severe accumulation of GM2 gangliosides in lysosomes, primarily within the central nervous system (CNS). This accumulation results in severe neurological impairment, lower motor neuron disease, and death. Currently, there are no effective therapies available for SD. Here, we explored the role of endoplasmic reticulum (ER) stress in the spinal cord during disease progression in an established mouse model of SD and revealed the beneficial outcome of off-label treatment with the FDA-approved drug, 4-phenylbutyric acid (4-PBA). We analyzed the expression and localization of ER stress and cellular apoptosis markers, which revealed significant upregulation of these factors within motor neurons. Additionally, we observed a > 50% reduction in neuronal numbers throughout all spinal cord regions. Our studies also tested the impact of the chemical chaperone 4-PBA on ER stress in mice, and following administration, we observed significant improvements in motor neuromuscular function and life span throughout disease progression. 4-PBA treatment significantly reduced apoptosis in spinal cord neurons and increased the number of choline acetyltransferase (ChAT)-positive neurons, with little effect on astrogliosis or sensory interneurons. Overall, this study provides strong evidence for the role of chronic ER stress in the pathophysiology of SD and highlights 4-PBA as a promising therapeutic treatment for SD and potentially other related LSDs.

Keywords: ER stress; Tay Sachs; lysosomal storage disease; lysosome; spinal cord.

Figure 1

Morphological abnormalities in spinal neurons of SD mice. A quantitative analysis of neuronal morphological changes seen in the spinal cord motor neurons (arrowhead) of Hexb^+/+ and Hexb^−/− mice during the pathogenesis of SD. Representative images of a single anterior horn motor neuron from Hexb^+/+ (A) and Hexb^−/− (B) spinal cords. Quantification of nuclear area (C) and cell body area (D) in Hexb^+/+ and Hexb^−/− spinal cord sections. Cell body area n = 20, nuclear area n = 20. ^*P < 0.05, ^**P < 0.0022 error bars ±SEM.

Figure 2

Mass neuronal loss throughout the spinal cord of SD mice at terminal disease stage. Representative immunohistological micrographs showing the neuronal-specific marker NeuN in the cervical region of the spinal cords of Hexb^+/+ (A and C) and Hexb^−/− (B and D) mice. (A and B) High-magnification images of anterior horn motor neurons of 40-d-old Hexb^+/+ and Hexb^−/− mice. Scale bar = 10 μm. (C and D) Low-magnification images of NeuN immunoreactivity in AHNs from 120-day-old Hexb^+/+ and Hexb^−/− mice. Scale bar = 100 μm.

Figure 3

Presence and localization of major UPR markers within anterior horn neurons. Representative immunohistological micrographs of the localization of the ER stress markers GRP78 and ATF6 within anterior horn motor neurons in the spinal cords of 60-, 80-, 100- and 120-day-old Hexb^+/+ and Hexb^−/− mice. Arrows indicate differential GRP78 localization to the cytosol and cell surface in Hexb^−/− AHNs. Scale bar = 10 μm.

Figure 4

Nuclear localization of downstream UPR players XBP1 and CHOP in the anterior horn neurons of SD mice. Representative immunohistological micrographs of the ER stress markers XBP1 and CHOP within the cervical spinal cord region of 60-, 80-, and 100-day-old Hexb^+/+ and Hexb^−/− mice. Scale bar = 10 μm.

Figure 5

Treatment with chemical chaperone, 4-PBA, negatively regulates ER stress activation. Representative immunofluorescent labeling of GRP78, ATF6, XBP1, CHOP and phosphorylated eukaryotic factor 2 alpha-positive (p-eIF2α) in the spinal cord from Hexb^−/− mice untreated or treated with 4-PBA. Note that Hexb^−/− AHNs showed mild reticular staining of GRP78 and ATF6, sparse nuclear staining of XBP1 and intense reticular staining of p-eIF2⍺. While 4-PBA Hexb^−/− AHNs showed increased GRP78, ATF6, and XBP1 staining, a shift in ATF6 from cytosolic to nuclear localization and a reduction in p-eIF2⍺ staining resulted in a significant reduction in the number of eIF2α-positive neurons with cytoplasmic staining after 4-PBA treatment. Bars represent 10 μm.

Figure 6

4-PBA successfully inhibits the nuclear localization of cleaved caspase 7 in anterior horn neurons of SD mice. (A) Quantification of the frequency of cleaved caspase 7-positive cells in the spinal cord of 60-, 80-, 100- and 120-day-old Hexb^+/+ and Hexb^−/− mice. (B) Representative double-label immunofluorescence micrographs of cleaved caspase 7-positive neurons and the neuronal marker NeuN in the spinal cord from Hexb^−/− mice untreated or treated with 4-PBA. Note the heavy nuclear localization of cCAS7 in untreated Hexb^−/− AHNs and a significant reduction in the number of cCAS7-positive AHNs in 4-PBA-treated Hexb^−/− AHNs. Bars represent 10 μm. (C) Quantification of the number of cleaved caspase 7-positive neurons in untreated and 4-PBA-treated Hexb^−/− spinal cords.

Figure 7

Administration of 4-PBA to SD mice results in retention of cholinergic motor neurons within the spinal cord. Representative immunofluorescence images of GFAP, ChAT and DAPI staining in the anterior horns of spine sections from Hexb^+/+, Hexb^−/−, and 4-PBA-treated Hexb^−/− mice. Note intense cytosolic ChAT staining in AHNs of Hexb^+/+ mice diminished ChAT staining in Hexb^−/− mice with active astrogliosis and intense cytosolic ChAT staining with active astrogliosis, as seen in 4-PBA-treated Hexb^−/− AHNs. Bar represents 10 μm.

Figure 8

The presence of calretinin-positive interneurons is regulated by 4-PBA. Representative immunofluorescent labeling of Calretinin-positive interneurons (sensory) and ChAT-positive neurons (motor) in spinal cord sections from Hexb^+/+, Hexb^−/−, and 4-PBA-treated Hexb^−/− mice. Note that ChAT-positive motor neurons were maintained in the anterior horn (AH, arrows) following 4-PBA treatment, while calretinin-positive interneurons densely stained the posterior horn (PH, arrowheads), with an increase in labeling intensity within the anterior horn of Hexb^−/− spinal cord mice. Bars represent 100 μm.

Figure 9

4-PBA improves neuronal and muscle fiber morphological abnormalities in SD mice. (A) Cell body sectional area of the anterior horn neurons and (B) the cross-sectional area of muscle fibers from the muscles surrounding the spinal cord of 80-day-old Hexb^+/+ and Hexb^−/− mice. Histological sections of spinal cord (C–E) and skeletal muscle (F–H) from Hexb^+/+, Hexb^−/−, and 4-PBA-treated Hexb^−/− mice. Within Hexb^+/+ mice, muscle fibers had an average area of ~ 1000 μm², while in contrast, Hexb^−/− mice showed a 44% reduction in area, with an average of ~ 570 μm². 4-PBA treatment of Hexb^−/− mice maintained normal cross-sectional areas of muscle myofibrils, which is consistent with functional cholinergic neurons.

Figure 10

Significant improvements in motor neuromuscular function and lifespan following treatment with 4-PBA. Motor neuron assessment of Hexb^+/+ and Hexb^−/− mice with and without 4-PBA. Mice underwent behavior testing as early as 7 weeks of age to measure degeneration of various cerebellar and peripheral nervous system functions. Measures of body mass (A), wire hang (B), righting reflex (C) and life span (D) were recorded over time. Measurements suggest that 4-PBA has a beneficial effect on SD. Hexb^−/− mice on 4-PBA live significantly longer than mice without 4-PBA and display a significantly faster righting reflex and longer wire hang during later stages of disease, indicating a positive effect on general motor functions. In addition, Hexb^−/− mice treated with 4-PBA have a similar increase in body mass as Hexb^+/+ mice. Analysis of survival by mantel-Cox test, P < 0.0001 between Hexb^−/− mice without 4-PBA and Hexb^−/− mice with 4-PBA, median survival = 128 days and 143 days, respectively.