Long-term impact of radiation on the stem cell and oligodendrocyte precursors in the brain

Abstract

Background: The cellular basis of long term radiation damage in the brain is not fully understood.

Methods and findings: We administered a dose of 25Gy to adult rat brains while shielding the olfactory bulbs. Quantitative analyses were serially performed on different brain regions over 15 months. Our data reveal an immediate and permanent suppression of SVZ proliferation and neurogenesis. The olfactory bulb demonstrates a transient but remarkable SVZ-independent ability for compensation and maintenance of the calretinin interneuron population. The oligodendrocyte compartment exhibits a complex pattern of limited proliferation of NG2 progenitors but steady loss of the oligodendroglial antigen O4. As of nine months post radiation, diffuse demyelination starts in all irradiated brains. Counts of capillary segments and length demonstrate significant loss one day post radiation but swift and persistent recovery of the vasculature up to 15 months post XRT. MRI imaging confirms loss of volume of the corpus callosum and early signs of demyelination at 12 months. Ultrastructural analysis demonstrates progressive degradation of myelin sheaths with axonal preservation. Areas of focal necrosis appear beyond 15 months and are preceded by widespread demyelination. Human white matter specimens obtained post-radiation confirm early loss of oligodendrocyte progenitors and delayed onset of myelin sheath fragmentation with preserved capillaries.

Conclusions: This study demonstrates that long term radiation injury is associated with irreversible damage to the neural stem cell compartment in the rodent SVZ and loss of oligodendrocyte precursor cells in both rodent and human brain. Delayed onset demyelination precedes focal necrosis and is likely due to the loss of oligodendrocyte precursors and the inability of the stem cell compartment to compensate for this loss.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Experimental set-up and definition of radiation field as demonstrated in a representative double-exposure X-ray of the skull of a rat secured in the radiation device (A).

The red circle indicates the opening in the radiation shield and the skull components that are exposed to radiation. MRI detectable gel polymer (IZI Medical Products, Baltimore MD) was placed over the area defined by the shield opening and sagittal MRI images performed (B, C) in order to demonstrate the brain volumes exposed to radiation. Red lines demonstrate the path of the radiation beam, at 5.7 degrees from the vertical. This is represented schematically in (D) to clearly illustrate the exclusion of the olfactory bulb and the distal most portion of the RMS from the radiation field. Stereological estimates of absolute BrdU counts in the SVZ, corpus callosum (CC) and cortex (Cx) (E–G) in irradiated (blue) and normal aging (red) rats. There is significant suppression of proliferation on day 1 post radiation in all 3 regions. BrdU levels are most significantly suppressed in the SVZ. In the cortex there is recovery to age-matched control levels at 3–6 months post radiation. In parallel, doublecortin labeled neuroblasts in the SVZ (H) are completely suppressed as of day 1 post radiation and do not recover up to 15 months post XRT. Stars indicate statistical significance (*** p

Figure 2. Effect of radiation on the olfactory bulb.

(A) Immunohistochemistry of BrdU/doublecortin (DCX) labeled cells shown in cross sections of the olfactory bulb at progressive time points post radiation. Quantitative measurements of total BrdU cells, BrdU/doublecortin in the distal RMS, doublecortin cells in the granular cell layer (DCX) and periglomerular calretinin-expressing interneurons are shown in B–E. Suppression of BrdU and DCX cells is delayed to 2 weeks post radiation but is followed by an immediate attempt at recovery peaking at 6 months post XRT. The origin of this recovery is thought to be due to proliferating neuroblasts in the distal RMS just proximal to the olfactory bulb. Concomitantly the SVZ and proximal RMS are devoid of proliferative activity and DCX expressing neuroblasts (F–G). Stars indicate statistical significance (*** p

Figure 3. Fluorescence intensity quantification of O4 and MBP and immunohistological assessment of oligodendrocyte markers.

(A)Representative contoured sections from the genu of the corpus callosum at serial time points post radiation immunostained for O4 (green). Quantitative measurements are shown in (B). A steady decline in O4 expression is seen immediately following administration of radiation until 3 months post XRT. At 6 months post radiation, a spike in O4 levels is followed by a significant decrease that persists until one year and thereafter. By 15 months, the majority of O4+cells are depleted. In comparison, aged control animals maintain a steady level of O4 until 18 months of age. (C) Serial immunohistochemical stains for MBP (red) on representative sections from the genu of the corpus callosum at various time points post radiation. Quantitative measurements are shown in (D). MBP expression is sustained until 6 months post radiation. By 9 months, patchy loss of myelin is observed throughout the corpus callosum. Demyelination is widespread by 15 months. Oligodendrocyte precursor markers, PDGFR (E) and NG2 (F) exhibit no significant change in intensity in aging animals but decrease rapidly after radiation without recovery, up to 15 months later. Markers of more mature oligodendrocytes such as CNP (G) exhibit a delayed decrease in expression starting at 8 months post XRT. CNP is significantly depleted at 15 months post XRT. DAPI in blue. (*** p

Figure 4. MRI imaging post radiation.

Representative sagittal (A, upper panel) and axial (A, lower panel) images of control or irradiated rats. At 12 months, subtle T2 changes are seen in the corpus callosum (arrows) that correspond to demyelination changes observed histologically. At 14 months post radiation, the T2 signal changes are more definite. (B) Graph depicts changes in the corpus callosum volume in irradiated animals as compared to aging control.

Figure 5. Endothelial cell number and capillary length post radiation.

(A) Representative images of sections from the corpus callosum immunostained for rat endothelial cell antigen (RECA) at various time points post radiation. RECA expression declines immediately post radiation but is restored and maintained through 15 months. Stereological estimates of the number of capillary segments in the cortex (B) and corpus callosum (C) and of capillary length in both regions (D, E). (*** p

Figure 6. Radiation effects in human tissue samples.

Human white matter samples acquired from non-irradiated brain (normal), irradiated specimens up to 7 months post XRT (labeled “early”), and irradiated specimens beyond 9 months up to 7 years (labeled “late”). Immunohistochemistry for early oligodendrocyte progenitor markers (O4, PDGFR), more mature oligoprogenitors (CNP) and fully differentiated oligodendrocytes (MBP) and endothelial cells (von Willebrand factor, vWF). Human tissues exhibit a pattern of early loss of young oligodendrocyte progenitors and delayed loss of more mature oligodendrocyte lineage cells, similar to what was described in the irradiated rat brain. Endothelial cells are scarce in early post radiation tissues and commonly normal in number in late post XRT tissues (7 years in this panel). DAPI in blue. Scale bar corresponds to 100 µm in all panels.

Figure 7. Ultrastructural features of irradiated rat and human brain tissue.

Electron microscopy of rat (upper panel) and human (lower panel) tissues in normal controls (A, D), an early/intermediate time point post radiation (B, E, 11 months and 7 months, respectively) and a late time point (15 months in the rat (C) and 7 years in the human (F)). Ultrastructural analysis of the myelin sheaths demonstrates normally compacted lamellae in the normal brains. At about 7 months post XRT, myelin sheaths in both human and rats (B and E and insets) acquire an irregular appearance with segmental loss of lamellar compaction associated with separation at the intraperiod line. These changes are more prevalent in larger myelin sheaths and are often mixed with normal myelinated fibers (arrows in (B) and (E)). Later times post XRT are associated with an increasing frequency and severity of myelin sheath degradation and vesiculation. Insets are magnifications of representative areas of myelin sheaths in each panel. Semithin toluidine sections of irradiated human tissue are shown in (G) and (H). Evidence of myelin sheath fragmentation is seen in the early/intermediate time point (14 months post XRT in (G)) as well as cytoplasmic lipid debris (white arrows) suggestive of active myelin degradation. Abnormal myelin sheaths persist and occur at higher frequency at late time points (7 years in H). Scale bar in (F) corresponds to 0.85 µm in (A), (D) and (E), 0.3 µm in (B) and (F) and 0.5 µm in (C). Scale bar in (H) corresponds to 10 µm in (G) and (H).