Time course of spinal doublecortin expression in developing rat and porcine spinal cord: implication in in vivo neural precursor grafting studies

Abstract

Expression of doublecortin (DCX), a 43-53 kDa microtubule binding protein, is frequently used as (i) an early neuronal marker to identify the stage of neuronal maturation of in vivo grafted neuronal precursors (NSCs), and (ii) a neuronal fate marker transiently expressed by immature neurons during development. Reliable identification of the origin of DCX-immunoreactive cells (i.e., host vs. graft) requires detailed spatial and temporal mapping of endogenous DCX expression at graft-targeted brain or spinal cord regions. Accordingly, in the present study, we analyzed (i) the time course of DCX expression in pre- and postnatal rat and porcine spinal cord, and (ii) the DCX expression in spinally grafted porcine-induced pluripotent stem cells (iPS)-derived NSCs and human embryonic stem cell (ES)-derived NSCs. In addition, complementary temporospatial GFAP expression study in porcine spinal cord was also performed. In 21-day-old rat fetuses, an intense DCX immunoreactivity distributed between the dorsal horn (DH) and ventral horn was seen and was still present in the DH neurons on postnatal day 20. In animals older than 8 weeks, no DCX immunoreactivity was seen at any spinal cord laminae. In contrast to rat, in porcine spinal cord (gestational period 113-114 days), DCX was only expressed during the pre-natal period (up to 100 days) but was no longer present in newborn piglets or in adult animals. Immunohistochemical analysis was confirmed with a comparable expression profile by western blot analysis. Contrary, the expression of porcine GFAP started within 70-80 days of the pre-natal period. Spinally grafted porcine iPS-NSCs and human ES-NSCs showed clear DCX expression at 3-4 weeks postgrafting. These data indicate that in spinal grafting studies which employ postnatal or adult porcine models, the expression of DCX can be used as a reliable marker of grafted neurons. In contrast, if grafted neurons are to be analyzed during the first 4 postnatal weeks in the rat spinal cord, additional markers or grafted cell-specific labeling techniques need to be employed to reliably identify grafted early postmitotic neurons and to differentiate the DCX expression from the neurons of the host.

Fig. 1

DCX expression in developing porcine spinal cord. A–A′″ Transverse spinal cord section taken from 40-day-old porcine fetus and stained with DCX and NeuN antibody. An intense DCX-IR in the majority of neurons in the dorsal and ventral horn can be seen (A–A′″; white boxes). Comparable DCX expression in the lateral funiculus was identified (A; yellow box). B–B″″, C–C′″ In 60-80-day-old fetuses, a progressive loss of DCX positive neurons in the ventral (C′) and central gray matter can be seen (B′–B′″, C′″). In contrast, an intense DCX expression continues to be present in the dorsal horn (B, B″″, C) and in a subpopulation of neurons localized in the medial regions of lamina VII (C″, C′″). D–D′ At 100 days (fetal period), only residual DCX positive neurons localized in laminae I–II can be identified (D). DCX immunoreactivity was absent in the deeper laminae (LV–LVII), in the ventral horn and in the white matter (D′). Scale bars: A = 1000 µm; A′, B, C, D, D′ = 500 µm; A″, B′, C″ = 200 µm; A′″, C′, C′″ = 100 µm. DH dorsal horn, VH ventral horn. E Immunohistochemical staining (brown DAB reaction) of doublecortin by goat polyclonal anti-DCX (sc-8066) antibody in 100-day-old fetal spinal cord. Lumbar spinal cord (a), lumbar spinal cord—dorsal horn detail (a′), lumbar spinal cord—adsorption control (a′). Thoracic spinal cord (b), thoracic spinal cord—dorsal horn detail (b′), thoracic spinal cord—adsorption control (b′). Cervical spinal cord (c), cervical spinal cord—dorsal horn detail (c′), cervical spinal cord—adsorption control (c′). Magnification x10 (a, a′, b, b′, c, c′), ×66 (b′, c′), ×100 (a′). E′ Rostro-caudal increase of dorsal horn DCX positivity at 100-day-old fetal spinal cord detected by sc-8066 antibody. F Immunohistochemical staining (brown DAB reaction) of doublecortin by rabbit polyclonal anti-DCX (#4604) antibody in 40 (a), 70 (b), 100 (c) days old fetal and 3 months old postnatal (d) lumbar pig spinal cord. 80-day-old fetal (e) and newborn (f) cortex were used as a positive controls. 40 and 70-day-old fetal spinal cord showed strong DCX expression. In 100-day-old fetal spinal cord, the maximum DCX immunopositivity was localized in dorsal horn and only weak unspecific signal was present in gray matter of 3 months old postnatal spinal cord. Magnification ×8 (a, b, c, d), ×20 (e), ×200 (f). G Immunohistochemical staining (brown) of doublecortin by mouse monoclonal anti-DCX (LS-C204512) antibody at 40 (a, a′), 70 (b, b′), 100 (c, c′, c′) days old fetal lumbar (a, b, c, c′) and thoracic (a′, b′, c′) pig spinal cord. DCX positivity in 100-day-old fetal spinal cord was only observed in dorsal horn (c′). Magnification ×10 (a, a′, b, b′, c, c′), ×66 (c′)

Fig. 2

DCX expression in developing rat spinal cord. A–A′ Transverse spinal cord section taken from 21-day-old rat fetus and stained with DCX and NeuN antibody. An intense DCX immunoreactivity distributed between the laminae I–V can be seen (A). At the base of the ventral horn occasional solitary DCX-IR interneurons can be identified (A′; yellow arrow). B–B″ At postnatal day 5, DCX immunoreactive neurons were identified primarily in the dorsal horn and in medial regions of laminae I–V (B, B′; white arrows). A comparable DCX expression in the corticospinal tract can also be seen (B; CST). In the ventral horn, the majority of NeuN+ neurons were DCX negative (B″; white arrows). C–C″ In 20-day-old rats, DCX immunoreactivity was confined to neurons localized in the superficial laminae of the dorsal horn (I–III) and to the corticospinal tract (CST corticospinal tract; white arrows). D–D″ Transverse spinal cord sections taken from 1-day-old rat pup and stained with GFAP and vimentin antibody. Mature GFAP+ astrocytes localized primarily at the periphery of the white matter can be identified (D). In contrast, dense networks of immature vimentin+ glial cells in the same region can be seen (D′, D″). E Transverse spinal cord sections taken from 3-month-old adult rat and stained with GFAP and NeuN antibody. A complete repopulation of the white and gray matter with mature GFAP+ astrocytes can be seen. Scale bars: A, C′ = 100 µm; B, C = 200 µm; B, B′ = 50 µm, C″ = 1500 µm, D-50 µm, E-200 µm. DH dorsal horn, CST corticospinal tract. F Western blot analysis of DCX expression in rat and porcine spinal cord. DCX expression in developing spinal cord at various fetal and postnatal developmental stages in pig and rat detected by western blot. DCX intensity was normalized to beta tubulin and plotted as a bar graph. Rat and pig fibroblasts were blotted together with pig NSCs to validate goat polyclonal anti-DCX (sc-8066) antibody specificity

Fig. 3

Immunohistochemistry of GFAP expression in developing porcine spinal cord. No GFAP (red) immunoreactivity in the lumbar spinal cord was seen in 40-day-old fetus (A–A′). At 80 days, a radial type of GFAP positive astrocytes localized in the subpial regions and extending GFAP+ processes toward deeper layers of the white matter can be seen (B, B′). In 100-day-old fetus and 1-day-old piglet, the white matter in dorsal and ventral funiculi showed an intense GFAP immunoreactivity with a clearly delineated GFAP+ network of astrocyte processes (C–D). In adult 5-month-old pig, a complete repopulation of white matter and gray matter with GFAP+ astrocytes can be seen (E). Scale bars A, B = 500 µm; C, D, E = 200 µm. DH dorsal horn, VH ventral horn, DF dorsal funiculus, CC central canal. F Western blot analysis of GFAP expression in porcine spinal cord. GFAP expression in developing spinal cord at various fetal and postnatal developmental stages in pig detected by western blot. Statistical analysis of three technical replicates per sample by 1-way Anova (p = 0.0013) with Bonferroni´s multiple comparison test confirmed that GFAP protein expression in spinal cord of 40-day-old fetus, 70-day-old fetus, and 1-day-old newborn is significantly different. Abbreviations: FC fetal cervical segment, FT fetal thoracic segment, FL fetal lumbar segment, 1PC first postnatal day cervical segment, 1PT first postnatal day thoracic segment, 1PL first postnatal day lumbar segment

Fig. 4

DCX expression in spinally grafted human ES-derived and porcine iPS-derived neural precursors. Intense DCX expression in human ES-NSCs grafted into the lumbar spinal cord of immunodeficient rats was seen at 3 weeks after grafting (A, B). The DCX expression was only seen in grafted human neurons expressing a human-specific nuclear protein (hNUMA), (A′, B′). A high density DCX positive network of neuronal dendrites can also be seen (A′, B″). A comparable level of DCX expression in spinally grafted porcine iPS-NSCs was seen at 4 weeks after grafting in continuously immunosuppressed pigs (C). The majority of grafted cells showed neuronal differentiation (NeuN) and extensive axo-dendritic arborization of DCX+ processes (C′, C″; white arrows) surrounding a large host α-motoneurons (C′, C″; white asterisks). Scale bars: A, B, C = 500 µm; A′, B′, B″ = 100 µm, C′, C″ = 50 µm. DH dorsal horn, VH ventral horn, CC central canal