Strain-specific differences in perinatal rodent oligodendrocyte lineage progression and its correlation with human

Abstract

Progress in the development of rat models of human periventricular white matter injury (WMI) has been hampered by uncertainty about the developmental window in different rodent strains that coincides with cerebral white matter development in human premature infants. To define strain-specific differences in rat cerebral white matter maturation, we analyzed oligodendrocyte (OL) lineage maturation between postnatal days (P)2 and P14 in three widely studied strains of rat: Sprague-Dawley, Long-Evans and Wistar (W). We previously reported that late OL progenitors (preOL) are the major vulnerable cell type in human periventricular WMI. Strain-specific differences in preOL maturation were found at P2, such that the W rat had the highest percentage and density of preOL relative to the other strains. Overall, at P2, the state of OL maturation was similar to preterm human cerebral white matter. However, by P5, all three strains displayed a similar magnitude and extent of OL maturation that persisted with progressive myelination between P7 and P14. PreOL were the predominant OL lineage stage present in the cerebral cortex through P14, and thus OL lineage maturation occurred latter than in white matter. The hippocampus also displayed a later onset of preOL maturation in all three strains, such that OL lineage maturation and early myelination was not observed to occur until about P14. This timing of preOL maturation in rat cortical gray matter coincided with a similar timing in human cerebral cortex, where preOL also predominated until at least 8 months after full-term birth. These studies support that strain-specific differences in OL lineage immaturity were present in the early perinatal period at about P2, and they define a narrow window of preterm equivalence with human that diminishes by P5. Later developmental onset of preOL maturation in both cerebral cortex and hippocampus coincides with an extended window of potential vulnerability of the OL lineage to hypoxia-ischemia in these gray matter regions.

Fig. 1

OL lineage progression in the developing white matter of 3 rat strains. Percentage of preOL (a) and immature OL (b) in SD (black bars), LE (dark gray bars) and W rats (light gray bars) at P2, P5 and P7. ** p < 0.01 (ANOVA; Tukey's multiple comparison test). Representative photomicrographs at P2 (c–e) and P5 (f–h) of preOL (red; arrowheads) and immature OL (yellow; arrows) in corpus callosum double-labeled with bO4 (red) and O1 (green). Scale bars = 40 μm. c, f SD rat. d, g LE rat. e, h W rat.

Fig. 2

Strain-specific differences in the density of preOL (a) and immature OL (b) in SD (black bars), LE (dark gray bars) and W rats (light gray bars) at P2 relative to P5 and P7. * p < 0.05; ** p < 0.01 (ANOVA; post hoc Tukey multiple comparison test). Representative photomicrographs at P2 (c–e) and P5 (f–h) of preOL (red) and immature OL (yellow) in corpus callosum double-labeled with bO4 (red) and O1 (green). Scale bars = 40 μm. c, f SD rat. d, g LE rat. e, h W rat.

Fig. 3

Myelination pattern visualized by MBP staining in W (a–c, f, i), SD (d, g) and LE rats (e, h). Shown are images at the midline corpus callosum (a) and external capsule (b, low power; c, high power) at P5, low-power coronal images at the level of the midseptal nuclei at P14 (d–f), and higher-power coronal images of the hippocampal formation and the adjacent white matter tracts of the corpus callosum (cc), external capsule (ec) and the fimbria of the fornix (fi) at P14 (g–i). Arrowheads: MBP-positive nonmyelinating OL. Arrows: MBP-positive myelinating OL. CA3 = Cornu ammonis area 3; DG = dentate gryus. Scale bars = 50 μm(a, c), 200 μm (b) and 500 μm (g–i).

Fig. 4

OL lineage progression in the cerebral cortex between P2 and P14. Low-power photomicrographs of the distribution of cells stained for bO4 (a–d) and O1 (e–h) at P2 (a, e), P5 (b, f), P7 (c–g) and P14 (d, h) in the SD rat. Individual somata are indicated by the red circles (online version only). Note the progressive increase in density of bO4-labeled cells within the full thickness of the cerebral cortex by P14. Relatively few O1-labeled cells are seen in e–h, supporting that the majority of the bO4-labeled cells in the cortex are preOL. Scale bar = 250 μm.

Fig. 5

OL lineage progression in the hippocampus. Low-power photomicrographs of the distribution of bO4-labeled cells and myelin at P7 (a) and P14 (b) in the LE rat. Higher-power images of double-immunohistochemical staining for bO4 (c, d) and O1 (e, f) as visualized in CA3 (box) are shown for P7 (c, e) and P14 (d, f). PreOL (arrowheads) labeled for bO4 but not O1. Immature OL (arrows) labeled for bO4 and O1. Scale bars = 1.0 mm(a, b) and 100 μm (c–f).

Fig. 6

OL maturation is less pronounced in human parietal cortex compared with white matter. Low-power photomicrographs show numerous cells labeled with bO4 antibody at 1 month (a) and 8 months (b) after full-term birth. Note that the bO4-labeled cells in the cortex were preOL that did not stain for the O1 antibody (not shown). a, b The approximate boundary between cortical layer 6 and the subcortical white matter is indicated by the arrowheads and was confirmed by a Hoechst fluorescent counterstain (not shown). c, d Higher-power details of the preOL (arrowheads) from the cases in a (c) and b (d) demonstrate the simplified arbor of processes associated with the cortical preOL at both ages. e Both preOL (red, online version only; arrowheads) and immature OL (yellow, online version only; arrows) were visualized in a region of early myelination at the edge of the densely myelinated tract shown in b. Scale bars = 200 μm(a, b) and 50 μm(c–e).