Delayed short-term tamoxifen treatment does not promote remyelination or neuron sparing after spinal cord injury

Abstract

The tamoxifen-dependent Cre/lox system in transgenic mice has become an important research tool across all scientific disciplines for manipulating gene expression in specific cell types. In these mouse models, Cre-recombination is not induced until tamoxifen is administered, which allows researchers to have temporal control of genetic modifications. Interestingly, tamoxifen has been identified as a potential therapy for spinal cord injury (SCI) and traumatic brain injury patients due to its neuroprotective properties. It is also reparative in that it stimulates oligodendrocyte differentiation and remyelination after toxin-induced demyelination. However, it is unknown whether tamoxifen is neuroprotective and neuroreparative when administration is delayed after SCI. To properly interpret data from transgenic mice in which tamoxifen treatment is delayed after SCI, it is necessary to identify the effects of tamoxifen alone on anatomical and functional recovery. In this study, female and male mice received a moderate mid-thoracic spinal cord contusion. Mice were then gavaged with corn oil or a high dose of tamoxifen from 19-22 days post-injury, and sacrificed 42 days post-injury. All mice underwent behavioral testing for the duration of the study, which revealed that tamoxifen treatment did not impact hindlimb motor recovery. Similarly, histological analyses revealed that tamoxifen had no effect on white matter sparing, total axon number, axon sprouting, glial reactivity, cell proliferation, oligodendrocyte number, or myelination, but tamoxifen did decrease the number of neurons in the dorsal and ventral horn. Semi-thin sections confirmed that axon demyelination and remyelination were unaffected by tamoxifen. Sex-specific responses to tamoxifen were also assessed, and there were no significant differences between female and male mice. These data suggest that delayed tamoxifen administration after SCI does not change functional recovery or improve tissue sparing in female or male mice.

Fig 1. Diagram of longitudinal spinal cord 2 mm rostral and caudal to the lesion epicenter demonstrating sample box placement during imaging.

The central canal is depicted as a dotted line running through the center of the cord; lesioned tissue is the red irregular shape; gray matter is the dark brown on either side of the center canal; white matter is the light brown on either side of the gray matter. Sample boxes were placed in the right and left white matter, with one box at 2 mm, 1 mm, and 0.5 mm rostral and caudal to the lesion epicenter.

Fig 2. Motor recovery is unaffected by delayed tamoxifen administration.

(A) Schematic of experimental timeline. (B-G) T indicates when animals received tamoxifen or corn oil gavage. (B) BMS scores were not different between sexes or tamoxifen and control mice at any time post-injury. (C) BMS subscores were significantly increased 14dpi in male mice compared to females, but sex did not impact subscores after gavage. BMS subscores were unaffected by tamoxifen. (D) The number of missed steps on the automated horizontal ladder was not statistically significant between groups. (E) Tamoxifen gavage reduced the number of ambulatory activities at 22dpi in female and male mice. Ambulation of tamoxifen mice recovered by 29dpi. (F-G) (F) The number of vertical rears or (G) speed of ambulation was not different between groups over time. (B-F) Two-way repeated measures ANOVA with Šídák post hoc test; corn oil female n = 9; tamoxifen female n = 10; corn oil male n = 8; tamoxifen male n = 9; *p<0.05.

Fig 3. Delayed tamoxifen treatment does not change the amount of white matter spared at the lesion epicenter, or distal to the lesion.

(A) Representative images of control and tamoxifen mice stained for eriochrome cyanine (EC) at the lesion epicenter, and 1 mm rostral and caudal to the epicenter. Scale bar = 100 μm. (B) Area of EC staining was not significantly different between control and tamoxifen mice or female and male mice at 1 mm rostral to the epicenter, at the epicenter, or 1 mm caudal to the epicenter. Data presented as mean + SEM. Two-way repeated measures ANOVA with Šídák post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 4. Neuronal survival after SCI does not change with delayed tamoxifen treatment.

(A-B) Representative images of NeuN labeling in the gray matter of the (A) dorsal horn and (B) ventral horn 1 mm rostral and caudal to the lesion epicenter after corn oil or tamoxifen gavage. Scale bar = 100 μm. (C-D) Tamoxifen mice had significantly fewer NeuN+ cells in the (C) dorsal horn (p = 0.0438) and the (D) ventral horn overall compared to control mice (p = 0.0431). Sex did not impact neuron number in the dorsal or ventral horn. Data presented as mean + SEM. Mixed-effects analysis with Šídák post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 5. Delayed tamoxifen treatment does not alter total axon area after SCI.

(A-B) Representative images of longitudinal spinal cord sections immunolabeled for (A) GAP-43 and (B) neurofilament-heavy (NF-H). Scale bar = 20 μm. (C-D) Percent area of (C) GAP-43 and (D) NF-H+ axons in the lateral white matter 2 mm, 1 mm, and 0.5 mm rostral and caudal to the lesion epicenter. No significant differences are reported between groups. Data presented as mean + SEM. Two-way repeated measures ANOVA with Šídák post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 6. Astrocyte reactivity and lesion size are not altered by delayed short-term tamoxifen gavage after SCI.

(A-B) Tiled confocal z-stacks of GFAP immunoreactivity in (A) control and (B) tamoxifen mice. Scale bar = 100 μm. Quantification of (C) lesion area and (D) GFAP proportional area revealed no significant differences between groups. Data presented as mean + SEM. One-way ANOVA with Tukey post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 7. Delayed short-term tamoxifen treatment does not alter cell proliferation.

(A, C, E) Representative confocal images of (A) proliferating cells (EdU/DAPI+ cells), (C) proliferating NG2 cells (NG2/EdU/DAPI+ cells), and (E) proliferating phagocytic macrophages/monocytes (CD68/EdU/DAPI+ cells) in the lateral white matter of control and tamoxifen mice. Scale bar = 75 μm. (B) Quantification of the total number of proliferating cells 2 mm, 1mm, and 0.5 mm rostral and caudal to the lesion epicenter. (D, F) Percent of proliferating (D) NG2 cells and (F) phagocytic macrophages/monocytes normalized to total proliferating cells at each sample distance rostral and caudal to the lesion. (B, D, F) The data are not significantly different between control and tamoxifen mice or female and male mice at any distance from the lesion epicenter. Data presented as mean + SEM. Mixed-effects analysis with Šídák post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 8. Oligodendrocyte number does not change after delayed short-term tamoxifen gavage.

(A) Representative confocal images of OLs (GSTπ/DAPI+ cells) in the lateral white matter. Scale bar = 75 μm. (B) Quantification of the total number of OLs 2 mm, 1mm, and 0.5 mm rostral and caudal to the lesion epicenter revealed no significant differences between groups. Data presented as mean + SEM. Two-way repeated measures ANOVA with Šídák post hoc test; corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6.

Fig 9. Demyelination and remyelination are unaffected by delayed short-term tamoxifen administration.

(A) Representative confocal images of longitudinal spinal cords immunolabeled for Kv1.2 and Caspr at 2 mm, 1 mm, and 0.5 mm rostral and caudal to the lesion epicenter. Inset shows a higher-power view of a node of Ranvier. Scale bar = 75 μm. (B) Percent of intact nodes of Ranvier in the lateral white matter at each sampled region. (C) Number of intact nodes of Ranvier normalized to percent axon area (Fig 4) in the lateral white matter at each sampled region. (D) Representative images of semi-thin longitudinal sections stained for toluidine blue from corn oil and tamoxifen mice at 2 mm, 1 mm, and 0.5 mm rostral and caudal to the lesion epicenter. Scale bar = 10 μm. (E) High-power image with arrowheads showing an example of a spared myelinated axon (black), demyelinated axon (pink), and remyelinated axon (teal). Scale bar = 10 μm. (F) Percent of spared myelinated, demyelinated, and remyelinated axons of total axons in each section. C/F = corn oil female; T/F = tamoxifen female; C/M = corn oil male; T/M = tamoxifen male. (B-C, F) The data are not significantly different between groups at any distance from the lesion epicenter. Data presented as mean + SEM. Mixed-effects analysis with Šídák post hoc test. (B-C) Corn oil female n = 6; tamoxifen female n = 7; corn oil male n = 5; tamoxifen male n = 6. (F) Corn oil female n = 1–3; tamoxifen female n = 1–3; corn oil male n = 3; tamoxifen male n = 3.