Systemic 5-fluorouracil treatment causes a syndrome of delayed myelin destruction in the central nervous system

Abstract

Background: Cancer treatment with a variety of chemotherapeutic agents often is associated with delayed adverse neurological consequences. Despite their clinical importance, almost nothing is known about the basis for such effects. It is not even known whether the occurrence of delayed adverse effects requires exposure to multiple chemotherapeutic agents, the presence of both chemotherapeutic agents and the body's own response to cancer, prolonged damage to the blood-brain barrier, inflammation or other such changes. Nor are there any animal models that could enable the study of this important problem.

Results: We found that clinically relevant concentrations of 5-fluorouracil (5-FU; a widely used chemotherapeutic agent) were toxic for both central nervous system (CNS) progenitor cells and non-dividing oligodendrocytes in vitro and in vivo. Short-term systemic administration of 5-FU caused both acute CNS damage and a syndrome of progressively worsening delayed damage to myelinated tracts of the CNS associated with altered transcriptional regulation in oligodendrocytes and extensive myelin pathology. Functional analysis also provided the first demonstration of delayed effects of chemotherapy on the latency of impulse conduction in the auditory system, offering the possibility of non-invasive analysis of myelin damage associated with cancer treatment.

Conclusions: Our studies demonstrate that systemic treatment with a single chemotherapeutic agent, 5-FU, is sufficient to cause a syndrome of delayed CNS damage and provide the first animal model of delayed damage to white-matter tracts of individuals treated with systemic chemotherapy. Unlike that caused by local irradiation, the degeneration caused by 5-FU treatment did not correlate with either chronic inflammation or extensive vascular damage and appears to represent a new class of delayed degenerative damage in the CNS.

Figure 1

CNS progenitor cells are vulnerable to clinically relevant levels of 5-FU exposure. (a) A summary of the putative relationships between the different cell types under study (for discussion of this and alternative views on lineage relationships in the CNS, see [199,200]). Pluripotent neuroepithelial stem cells (NSC) give rise to glial restricted precursor (GRP) cells and neuron restricted precursor (NRP) cells. GRP cells in turn give rise to astrocytes and oligodendrocyte-type-2 astrocyte progenitor/oligodendrocyte precursor cells (O-2A/OPCs), the ancestors of oligodendrocytes. (b,c) Primary CNS cells (b) or various cancer cell lines (c) were grown on coverslips and exposed to 5-FU for 24 h before analysis of cell viability as described in Materials and methods. 5-FU concentrations were chosen on the basis of drug concentrations reached in humans after conventional 5-FU treatment. None of the tumor lines tested were sensitive to 5-FU treatment in this dose range, whereas O-2A/OPCs, oligodendrocytes, GRP cells and human umbilical vein endothelial cells (HUVECs) were sensitive. (d,e) Exposure conditions designed to mimic the exposure levels associated with long-term infusion (d) or high-dose bolus administration (e) yielded similar results, with vulnerability of O-2A/OPCs and non-dividing oligodendrocytes to 5-FU exceeding the vulnerability of rapidly dividing cancer cells. As shown in (b,d), the vulnerability of HUVECs also exceeds the vulnerability of cancer cells. Each experiment was carried out in quadruplicate and was repeated at least twice in independent experiments. Data represent mean of survival ± s.e.m, normalized to control values.

Figure 2

Sublethal doses of 5-FU inhibit division of O-2A/OPCs. Clonal analysis was used to study the effects of low-dose 5-FU (0.05 μM for 24 h) on the division and differentiation of freshly isolated progenitor cells. O-2A/OPCs were grown at clonal density and exposed one day after plating to (a) vehicle alone or (b) 0.05 μM 5-FU for 24 h, doses that killed less than 5% of O-2A/OPCs in mass culture. The number of undifferentiated O-2A/OPCs and differentiated cells (oligodendrocytes) was determined in each individual clone from a total number of 100 clones in each condition by morphological examination and by immunostaining with A2B5 and anti-GalC antibodies (to label O-2A/OPCs and oligodendrocytes, respectively). Results are presented as three-dimensional graphs. The number of progenitors per clone is shown on the x (horizontal) axis, the number of oligodendrocytes on the z (orthogonal) axis and the number of clones with any particular composition on the y (vertical) axis. In 5-FU-treated cultures analyzed five days after initiating 5-FU exposure, there was an increase in the representation of small clones consisting wholly of oligodendrocytes and clones containing large numbers of oligodendrocytes, a reduction in the representation of large clones, a general shift of clone size towards smaller values, and a clear reduction in the total number of progenitor cells (see text for details). Experiments were performed in triplicate in at least two independent experiments.

Figure 3

Systemic 5-FU treatment causes cell death in the adult CNS. Cell death was determined using the terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) assay. The number of TUNEL⁺ cells was analyzed in control animals (that received 0.9% NaCl i.p.) and 5-FU-treated animals and presented as percentage normalized values of controls at each time point. For ease of comparison, data presented in the figures show the control value (mean set at 100% of the day 1 value) and normalized values of 5-FU treatment groups at all time points. Each treatment group and the control group consisted of n = 5 animals at each time point. Figures show apoptosis in animals that received three bolus i.p. injections of 5-FU (40 mg kg^-1 on days -4, -2 and 0 leading up to the analysis, where day 1 of analysis equals 1 day after the last treatment with 5-FU). There was marked and prolonged increase of cell death in the 5-FU treatment group in (a) the lateral subventricular zone (SVZ), (b) the corpus callosum (CC) and (c) the dentate gyrus (DG) at 1, 7, 14 and 56 days and 6 months following treatment. Data are means ± s.e.m.; a two-way ANOVA test was performed on the original un-normalized data set to test the statistical significance of treatment effect and time effect. Bonferroni post-tests were performed to compare the 5-FU-treated group and the control group at each time point. The statistical significance of the Bonferroni post-tests is labeled in the graphs where applicable: ***p < 0.001; **p < 0.01; and *p < 0.05. Two-way ANOVA test results indicate that, in the SVZ, the treatment effect is extremely significant (p < 0.001), the time effect is very significant (p < 0.01); in the CC, the treatment effect is not quite significant (p = 0.06), the time effect is not significant (p = 0.74); in the DG, the treatment effect is extremely significant (p < 0.001), the time effect is significant (p < 0.05). The effect of the interaction between treatment and time is not significant for all three regions. (d) To determine the immediate cellular targets of 5-FU in vivo, we examined co-analysis of TUNEL labeling with antigen expression in animals sacrificed at day 1 after completion of 5-FU treatment. The majority of TUNEL⁺ cells in the SVZ and DG were doublecortin (DCX)⁺ neuronal progenitors. Other TUNEL⁺ cells in these two regions included GFAP⁺ cells (which could be stem cells in the SVZ, or astrocytes in the DG) and Olig2⁺ O-2A/OPCs. There was also a small contribution of NeuN⁺ mature neurons in the DG. In the CC, the majority of TUNEL⁺ cells were Olig2⁺ (which, in this white matter tract, would be oligodendrocytes and O-2A/OPCs), with a small contribution of GFAP⁺ astrocytes. Almost 100% of TUNEL⁺ cells were accounted for by known lineage markers. Each group consisted of n = 4 animals. Data are mean ± s.e.m.

Figure 4

Systemic 5-FU exposure causes prolonged suppression of proliferation in the adult CNS. Animals were treated as described in Figure 3. The number of BrdU⁺ cells was analyzed in control animals and 5-FU-treated animals and presented as percentage normalized values of controls at each time point. For ease of comparison, data presented in the figures show the control value (mean set at 100%) of day 1 and normalized values of 5-FU treatment groups at all time points. Each group consisted of n = 5; a two-way ANOVA test was performed on the original un-normalized data set to test the statistical significance of treatment effect and time effect. Bonferroni post-tests were performed to compare the 5-FU-treated group and the control group at each time point. The statistical significance of the Bonferroni post-tests is labeled in the graphs where applicable: ***p < 0.001; **p < 0.01; or *p < 0.05. Two-way ANOVA test results indicate that: (a) in the SVZ, both the treatment effect and time effect are extremely significant (p < 0.001), and the interaction of treatment and time is very significant (p < 0.01); (b) in the CC, both the treatment effect and time effect are extremely significant (p < 0.001), and the interaction of treatment and time is very significant (p < 0.01); and (c) in the DG, the treatment effect is very significant (p < 0.01), the time effect is extremely significant (p < 0.001), and the effect of the interaction between treatment and time is not significant.

Figure 5

Cell-type analyses of BrdU⁺ cells in control and 5-FU-treated animals at early and late time points after completion of treatment. Co-analysis of BrdU incorporation with antigen expression was conducted as described in Materials and methods. Both control and 5-FU-treated groups were analyzed at (a,b) day 1 and (c,d) day 56 to evaluate the immediate and long-term effects of 5-FU treatment. Results indicate that division of both DCX⁺ neuronal progenitors and Olig2⁺oligodendrocyte precursors was reduced by systemic exposure to 5-FU. In the CC, the reduction in apparent division of Olig2⁺ cells was proportionate to the overall reduction in all BrdU⁺ cells. In the SVZ, there was an enhanced reduction of DCX⁺ cells from among the BrdU⁺population at day 1 but not at day 56. In the DG, there was an enhanced reduction in the dividing DCX⁺ population at both day 1 and day 56. In addition, the proportion of GFAP⁺ cells in the CC was increased among the BrdU⁺ population at both time points examined. Data are mean ± s.e.m; *p < 0.05, in comparisons with control animals (confidence interval = 95%, by unpaired, two-tailed Student's t-test).

Figure 6

Systemic 5-FU treatment caused delayed increases in auditory brainstem response (ABR) inter-peak latencies P2-P1 and P3-P1. Baseline ABR hearing tests were performed on each animal one day before initiation of treatment with 5-FU (as for Figure 4). After treatment ended, follow-up ABR tests were conducted on each animal at various points during a time course of 56 days. Control and treatment groups both consisted of n = 4 animals. ABR latencies were analyzed for each individual at each time point, and change of latency was calculated as L_t – L₀ (L_t, latency values at day 1, day 7, day 14, or day 56 post treatment; L₀, baseline latency values 1 day before treatment initiation). (a) The change of inter-peak P2-P1 latency values; (b) the change of inter-peak P3-P1 latency values. At the later time points day 14 and day 56, both P2-P1 and P3-P1 inter-peak latency values of 5-FU-treated animals show average increases of more than 0.13 ms, whereas the same inter-peak latency values of sham-treated controls show average decreases or an increase of less than 0.04 ms. Data are mean ± s.e.m. Statistical significance of the difference between the means of control and treated groups was p < 0.05 in (a), and p < 0.01 in (b) (confidence interval = 95%; paired, one-tailed Student's t-test).

Figure 7

Systemic 5-FU treatment causes delayed dysregulation of Olig2 expression in oligodendrocytes in the CC. Animals were treated with 5-FU as in Figure 3 and analyzed for expression of Olig2 in the CC at various time points. There was a marked reduction in the number of such cells at 56 days ((a) control; (b) 5-FU) after completion of treatment, but not at 1 or 14 days after treatment. (c) Percent-corrected number of Olig2⁺ cells in the CC at day 1 and day 56 post-treatment with 5-FU, normalized to control values at each time point. Data represent averages from three animals in each group, shown as mean ± s.e.m (*p < 0.001, one-way ANOVA) in comparison with control values at each time point. The scale bar represents 150 μm. (d-i) Representative confocal micrographs showing loss of Olig2 expression in a subset of CC1⁺ oligodendrocytes in the CC of a 5-FU-treated animal at day 56 in comparison with a sham-treated animal at the same time point. The reduction in numbers of Olig2⁺ cells seen at day 56 after treatment was not associated with an equivalent fall in oligodendrocyte numbers, as determined by analysis of CC1⁺ expression. (d-f) In control animals, there is a close equivalence between CC1 expression (d) and Olig2 expression (e); a merged image is shown in (f). Three Olig2⁺ CC1^- cells can be seen in (e,f) (arrowheads), which are probably O-2A/OPCs. (g-i) In contrast, in 5-FU-treated animals there is a reduction in the number of Olig2⁺ cells (h), but the CC of these animals contains many CC1⁺ cells (g) that do not express Olig2 (i) (arrows, Olig2⁺ CC1⁺; arrowheads, Olig2⁺ CC1^- cells). The scale bar represents 25 μm.

Figure 8

Delayed myelin and axonal degeneration in the CC caused by systemic 5-FU treatment (representative electron micrographs of longitudinal sections of axons). Sections were taken from midline coronal sections of the CC. (a) A representative image from a sham-treated control animal, showing normal myelinated axons and the normal axonal cytoskeleton structures; (b-f) representative images from a 5-FU-treated animal, showing several pathological changes of both the myelin and axonal structures. Asterisks, axonal abnormality; single arrows, damaged myelin sheaths; double arrows, myelin debris; arrowheads, engulfed myelin debris. (b) Several swollen axons with disrupted cytoskeleton (asterisks), damaged myelin sheaths (single arrows) and myelin debris (double arrows) can be seen. (c) Several swollen axons (asterisks) with or without myelin can be seen, the axoplasm of which show disruption of cytoskeleton and altered organelles. (d) Several axons (asterisks) with absent or degenerating myelin (arrows) can be seen; one axon shows a severely damaged axonal structure and myelin on one side of a node of Ranvier (n) and partially disrupted myelin sheath on the other side (arrow). (e) Several loci of myelin degeneration can be seen (arrows); one axon seems to be transected on one side of a node of Ranvier (n). An axon next to it shows partial degeneration of the myelin sheath and disruption of the cytoskeleton (asterisk). (f) Edema in what is likely to be a process of an astrocyte can be seen, with some engulfed myelin debris (arrowhead) and the adjacent axons are distorted; there are also swollen axons (asterisks) with and without myelin (arrows).

Figure 9

Ultrastructural evidence of myelinopathy in 5-FU-treated animals. Electron micrographs were taken from the midline transverse sections of the CC (cross-sections of the axons). (a) A representative image from a sham-treated control animal, showing normal myelinated axons; (b-f) representative images from a 5-FU-treated animal, showing multiple pathological changes of both the myelin and axonal structures. Single asterisks indicate demyelinated axons with rarefaction (that is, decreased density of the axoplasm staining possibly due to disruptions in cytoskeletal structures and organelles); double asterisks indicate an abnormal axon with partially destructed myelin sheaths; single arrows indicate inter-laminar splitting of the myelin sheaths; and double arrows indicate myelin debris. (b) Two axons with damaged myelin sheaths (asterisks), myelin debris (double arrows) and a smaller axon that seems to be detaching from its myelin sheath (single arrow) can be seen. (c) A large demyelinated axon with rarefaction of the axoplasm (asterisk) and two axons with collapsed centers and inter-laminar splitting of the myelin sheaths (arrows) can be seen, indicating on-going myelin degeneration. (d) Two large axons with completely (asterisk) or partially (double asterisks) damaged myelin can be seen, the axoplasm of which shows altered cytoskeleton and organelles. One axon has a collapsed center and inter-laminar splitting (arrow). (e) Myelin debris can be seen, possibly from a degenerating axon (double arrows) and an axon with inter-laminar splitting (arrow). (f) A demyelinated axon with rarefaction of the axoplasm and possible axonal swelling (asterisk) and two neighboring axons with inter-laminar splitting (arrows) can be seen.

Figure 10

5-FU treatment causes reduced cellularity and loss of myelin basic protein (MBP) at 6 months after treatment. Representative images of hematoxylin and eosin staining from the periventricular region of (a,c) a control animal and (b,d) a 5-FU-treated animal. (c) Partial enlargement of the CC shown in (a); (d) partial enlargement of the CC shown in (b). (a) Normal cellular density is seen in the CC of the control; (b) in the CC from a 5-FU-treated animal, the cellular density in the CC has decreased markedly. (e,f) The expression of MBP seen in control animals (e) (the fiber-like green fluorescence staining in the CC and white-matter tracts in the peri-ventricular striatum) is greatly reduced in treated animals (f). The bright green punctuated fluorescent staining is BrdU⁺ cells, which are present in control animals but greatly depleted in treated animals. All sections were processed at the same time and all images were taken under equal exposure times. The scale bar represents 100 μm.

Figure 11

5-FU induces transient inflammation and apoptosis of microvasculature endothelial cells in a subset of treated animals. Representative photographs showing the inflammatory response on day 1 after treatment with (a) vehicle or (b) 5-FU treatment as indicated by immunostaining for the activated microglia/macrophage marker F4/80. The basal level of F4/80 staining was very low in the controls but increased after treatment. These sections are from the CC (the region between the two dotted lines), with similar evidence of inflammation seen in the DG and cortex of this same mouse. The scale bar represents 50 μm. (c-e) TUNEL/PECAM (CD31) double immunostaining was performed in 5-FU-treated animals, with representative images taken from the DG showing double-labeling of the vascular endothelial cell marker PECAM (CD31) with TUNEL⁺ nuclei. In the subset of animals in which evidence of endothelial cell death was observed, similar TUNEL⁺ profiles were also found in the cortex and CC.