Oligodendrocytes in Amyotrophic Lateral Sclerosis

Image Unavailable
An oligodendrocyte wrapping three different neuronal axons

Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease which causes death of upper and lower motor neurons, leading to a progressive loss of motor functions [1]. The degeneration of motor neurons has been linked with the dysfunction of neuroglial cells such as oligodendrocytes. Oligodendrocytes main function is to provide support and insulation to the axons in the central nervous system (CNS), by creating the myelin sheat. They have been found to provide metabolic support to the neurons by supplying energy metabolites such as glucose and lactate [2]. Since the most abundant transporter of lactate in oligodendrocytes is monocarboxylate transporter 1 (MCT1), their disruption results in axonal damage and therefore neural loss. It has been found that oligodendrocyte progenitors (NG2+ cells) have an enhanced proliferation and differentiation, but even though new oligodendrocytes are being formed, they fail to mature, resulting in a progressive demyelination of axons. Experiments in mouse models in which the mutated gene that results in ALS, SOD1 (G37R) was selectively removed, lead to a delay in the disease onset and increased the survival of mice. This proposes that mutations are sufficient to produce death of motor neurons [3].

1. Oligodendrocyte progenitors (NG2+ cells)

Oligodendrocyte progenitors (NG2+) cells are a type of glia cells in the central nervous system. Once they differentiate into mature oligodendrocytes, they provide axonal support and electrical insulation by forming the myelin sheath. The loss or damage of oligodendrocytes lead to the loss of myelin sheath and can lead to degeneration of axons. The differentiation of NG2+ cells into mature oligodendrocytes can be detected by the expression of myelin basic protein (MBP), proteolipid protein (PLP), or myelin-associated glycoprotein (MAG)[3].

1.1 Genetic fate tracing of oligodendrocytes

Image Unavailable
WT and SOD1 (G93A) fluorescence images
of ventral grey matter BrdU+ cells.

Fig. 1b
Image Unavailable
Expression of cleaved caspase-3 (g) in CC1+
oligodendrocytes (h). Capase-3 and CC1+ overlap (i)

When motor neurons are lost in SOD1 (G93A) mice, there are changes in the behavior of NG2+ cells. NG2+ cells proliferate at about 20 fold faster than control mice by end stage of disease. To study how NG2+ cells proliferate throughout the disease, Kang et al [3] gave mice BrdU, which is a marker for proliferating cells, and measured the overall amount after 5 days, in the lumbar region of spinal cord. In control mice, the amount of NG2+ cells decreased with age, while in SOD1 (G93A) mice, NG2+ cells continued to proliferate into adulthood. There was high proliferating rate in ventral gray matter before the disease onset and even a higher rate as the disease progressed. These results suggest that the behavior of NG2+ cells is being altered even before the onset of the disease, but even though there is an increase in NG2+ cells, their density is not enhanced before the end stage, due to the fact that old oligodendrocytes are dying and being replaced during early disease. Fig. 1a

Furthermore, to determine the fate of NG2+ cells at early stage of disease, genetic lineage tracing was performed by giving 4-Hydroxytamoxifen (4-HT) to mice, to induce EGFP expression. Ninety days later there were more EGFP+ cells in the ventral gray matter of SOD1 (G93A) mice than in control. By the end stage, the density of EGFP+ cells in SOD1 (G93A) mice was even higher than control mice. This demonstrates that NG2+ cells do proliferate in SOD1 (G93A) mice [4]. Fig. 1b

1.1.1 CC1+ positive oligodendrocyte

Fig. 2g-i
Image Unavailable
Expression of cleaved caspase-3 (g) in CC1+ oligodendrocytes (h).
Capase-3 and CC1+ overlap (i)

NG2+ cells proliferate and differentiate more often into oligodendrocytes in the ventral grey matter of SOD1 (G93A) mice, by end stage of disease. To determine which NG2+ cells go on differentiating into mature oligodendrocytes, Kang et al [3], followed EGFP+ cells which were inmmunoreactive to antibody to APC (CC1), a marker of differentiated oligodendrocytes, in SOD1 (G93A) and control mice. In control mice there were few EGFP+ CC1+ cells, while there was a significant increase of newly generated oligodendrocytes in SOD1 (G93A) mice ventral grey matter. Even though there is an increase in oligodendrocytes in SOD1 mice, their density is the same as in control mice, which suggest there must be some death of old oligodendrocytes as the disease progresses.
Philips et al [1], also examined CC1 oligodendrocytes in ventral grey matter and noticed morphological changes in SOD1 (G93A) mice, which increased in number as the disease progressed. They went further into investigating if these morphological changes, such as the thickening of cell body and the elongated reactive morphology, could predict the death of these oligodendrocytes. They performed double immunolabelling, CC1 and anti-cleaved caspase-3 (a marker for apoptotic cell death). In the spinal cord of SOD1 (G93A) mice, cleaved caspase-3 was present in CC1+ oligodendrocytes, but was not present in control mice. This demonstrated that oligodendrocytes are indeed involved in the disease progress. Fig. 2g-i

Fig. 3
Image Unavailable
Percentage of labeled oligodendrocytes
after 4-HT administration in ventral grey
matter. Closed circle: control group.
Open circle: SOD1 (G93A) mice.

Also, they followed oligodendrocytes as the disease progressed, to demonstrate that oligodendrocytes in SOD1 (G93A) spinal cord increased their turnover. PLP-YFP- SOD1 (G93A) transgenic mice were used. Proteolipid Protein (PLP) is a structural protein of myelin sheats and will be present only in mature oligodendrocytes. After treatment with tamoxifen, YFP in CC1+ was found in gray and white matter oligodendrocytes, but was not detected in NG2+ cells. Therefore CC1+ oligodendrocytes which are YFP- (negative) are the new oligodendrocytes from NG2+, which suggest there is an increase in the rate of production of new oligodendrocytes in PLP-YFP- SOD1 (G93A) mice.

To determine oligodendrocyte survival as the disease progressed, genetic tracing was performed in PLP-creER, ROSA26-EYFP, SOD1 (G93A) mice [3]. 4-HT was administered and the number of EYFP+ oligodendrocytes in the spinal cord at 40 and 70 days after labelling were counted. In SOD1 (G93A) mice, EYFP+ oligodendrocytes were reduced by 20% at 40 days, and by end stage disease there was a further decrease that resulted in 65% reduction of the initial labelled oligodendrocytes. This demonstrated there is prominent loss of early born oligodendrocytes in SOD1 (G93A) mice ventral grey matter near motor neurons, as mice show signs of disease. Fig. 3

1.1.2 Oligodendrocyte survival

As oligodendrocytes degenerate in the spinal cord, some EGFP oligodendrocytes expressed active caspase-3 at end stage mice and the formation of EGFP+ clusters and active microglia; which are attracted by apoptotic cells. This shows that oligodendrocytes are dying via apoptotic cell death.
They wanted to know if the loss of motor neurons was enough to recruit NG2+ cells proliferation and differentiation. They proceeded into ablating motor neurons, but NG2+ cells did not proliferate, suggesting that the loss of motor neurons is not enough to recruit NG2+ cells [3].

1.1.3 Oligodendrocyte morphology

Fig. 4 a,b
Image Unavailable
a. Confocal images of EFGF+ of control and SOD1 (G93A) in
ventral grey matter. b. Magnified merged region

To determine when oligodendrocytes start to show abnormalities, their morphology was examined in Mobp-EGFP, SOD1 (G93A) mice. In control mice, oligodendrocyte somata (EGFP+ Olig2+) in spinal cord has a nice oval shape, while SOD1 (G93A) mice at early and end stage have an irregularly shaped EGFP+ structures, lack nuclei and look like the fragments that result from apoptotic cell death. In spite of these differences EGFP+ Olig2+ density was the same as control mice. This demonstrates that pathological changes in oligodendrocytes is widespread and occur when mice start to show behavioral manifestations of the disease [3]. Fig. 4 a,b

Similarly, Philips et al [1] induced expression of YFP in NG2+ cells and their mature progeny in PDGFRa-YFP- SOD1 mice and control mice (PDGFRa-YFP). In the control group, about 20% of NG2+ cells were labelled [5], but there was a higher proliferation of NG2+ cells in SOD1 (G93A) mice. Also YFP+ CC1+ expression in SOD1 (G93A) mice was double than control, but most of these newly differentiated oligodendrocytes had altered morphology. Therefore this demonstrates that as the disease progresses, NG2+ cells generate new oligodendrocytes but they are dysmorphic.

1.1.4 Dysfunctional oligodendrocytes

Fig. 5b
Image Unavailable
MAG expression levels at different stages
of disease in SOD1(G93A) mice

Image Unavailable
MBP expression levels at different stages
of disease in SOD1(G93A) mice

The expression of myelin basic protein (MBP), myelin associated glycoprotein (MAG) and monocarboxylate transporter 1 (MCT1), was examined to investigate if the decrease in metabolic support and the increased rate of oligodendrocyte contributed to degeneration of motor neurons in SOD1 (G93A) mice. MCT1 and MBP were reduced in SOD1 (G93A) mice, and even more decrease in MBP expression was seen in the ventral horn gray and white matter, suggesting there may be a defect in gray matter oligodendrocytes [1][2][6]. Fig. 5 b, e

NG2+ cells proliferate and differentiate at higher rates when there is loss of oligodendrocytes. Data shows that oligodendrocytes degenerate in the spinal cord of SOD1 (G93A) mice. These mice present oligodendrocytes with cleaved caspase-3, suggesting these cells become apoptotic. There is increased generation of oligodendrocytes from NG2+ cells, but these cells are dysmorphic, and the expression of mature oligodendrocytes markers does not mean that the cell is functional, as with the expression of CC1+, myelin formation was compromised because there was a decrease in MBP.

2. Monocarboxylate Transporter 1 (MCT1) down regulation

Oligodendrocytes support axon survival by providing them with the right amount of nutrients, through monocarboxylate transporter 1 (MCT1). MCT1 is the main transporter of lactate, which provides the metabolites necessary for axon survival. The disruption of this transporter leads to axon damage and neuron loss. In ALS patients and mice models of ALS, it has been found that there is down regulation of MCT1 transporter [2].

2.1 Spinal cord of SOD1 mice and motor cortex of ALS patients

Fig. 6 a,b
Image Unavailable
a.photomicrographs and b. quantification of motorneurons treated with control (Ctrl),
MCT1 sense oligonucleotide (sense) or MCT1 ASO (ASO) in spinal cord slice cultures.

MCT1 can be present on astrocytes, endothelial cells, ependymocytes and ologodendrocytes, since there is high variability; Lee et al.[2], wanted to demonstrate that MCT1 are indeed expressed in oligodendrocytes. Using transgenic mice, they used tdTomato fluorescent reporter for the localization and in vivo expression level of MCT1 messenger RNA. MCT1 was localized in oligodendrocytes of brain and spinal cord, suggesting MCT1 expression and the transport of lactate is much higher in oligodendrocytes than in astrocytes.
Organotypic spinal cord cultures (in vitro) were treated with antisense oligonucleotides (ASO) or MCT1 transport inhibitor (MCT1i) [7], to see if the down regulation or the inhibition of MCT1 causes neuronal cell death. After the treatment, in the ASO group about 67%, and in MCT1i about 66% of motor neurons survived. This demonstrates that the reduction of MCT1 leads to motor neuron death. Fig. 6 a, b

Neurons deprived of glucose could survive for about 2 hours, in vitro. MCT1i and glucose deprivation lead to even more neuron death in the ventral horn of SOD1 (G93A) mice. Motor neuron loss could be prevented by supplying 20mM L-lactate, suggesting that death of motor neurons in culture was due to the reduction of lactate and not due to the inhibition of MCT1 transporters [2].
It was found that the complete absence of MCT1 embryonically is lethal, while the heterozygous-null mice (MCT1+/-) can survive without abnormalities. Eight months later however, the MCT1+/- mice presented axonopathy in the CNS without demyelination or oligodendrocyte injury, demonstrating that MCT1 through a myelin independent mechanism is important for the normal function of CNS axons. The axon pathology was similar to Cnp-null mice [8], to patients with ALS [9], and to SOD1 (G93A) transgenic mice [10].

Fig. 7 m, n
Image Unavailable
Mean number and scatter plot of degenerative axons in the optic nerve
of lenti-GFP and lenti-shRNA

Since it is known that not only oligodendrocytes contain MCT1, studies were carried to see if the presence on MCT1 in other cells were influencing the down regulation of MCT1 as seen in MCT1+/- mice. They produced two different lentiviral constructs to down regulate MCT1 only in oligodendrocytes. One of them was Lenti-MBP-shRNA which efficiently down regulated MCT1 protein in the optic nerve. It was noticed that there was an increased degeneration of axons distal to the site of injection and an increase in the number of degenerating fibres in the optic nerve in lenti-MBP-shRNA than in lenti-BMP-GFP (control). This showed that in vivo oligodedrocytes specific down regulation of MCT1 in optic nerve was able to produce degeneration of axons. Since the loss of MCT1 and MCT1+/- mice experiment cause axon degeneration, MCT1 is necessary for axon survival [2]. Fig. 7 m-n

Fig. 8 d-g
Image Unavailable
Immunofluorescence of MCT1 and (e, g)
double labeling of MCT1 and CC1+ in end
stage SOD1 and control mice

Alterations of MCT1 in oligodendrocytes in white and grey matter, needed for the metabolic support of motor neurons, may contribute to the pathogenesis seen in ALS in vivo and in vitro [4]. The levels of expression of MCT1 proteins and MCT-based protein CD147 were studied in affected (motor cortex) and unaffected (frontal cortex) regions in ALS patients and control patients. Also the activity of MCT1 was studied in the spinal cord of SOD1 (G93A) mice. Motor cortex of ALS patients had a 50% decrease in the expression of MCT1 than control patients. There was not reduction in the frontal cortex and in CD147. Also it was noticed that in SOD1 (G93A) mice there was down regulation of MCT1 mRNA in the spinal cord of early and end stage disease mice, especially in the ventral horn grey matter. CC1+ oligodendrocytes were also present, which suggested that MCT1 expression is not due to decrease in oligodendrocytes. Fig. 8 d-g

To investigate if the mutation of SOD1 decreased the levels of MCT1 at the transcriptional level or post-transcriptional level, Philips et al [1] using quantitative PCR quantified MCT1 messenger RNA expression. SOD1 (WT), SOD1 (G93A), and SOD1 (A4V) were co-transfected with MCT1-FLAG. The results did not show any difference in MCT1 m-RNA levels. Fig. 9

Therefore the effect of SOD1 (G93A) or SOD1 (A4V) on MCT1 protein levels is post-transcriptional, and the differences seen in MCT1 protein levels are not due to unequal co-transfection. This was confirmed in vivo by quantifying m-RNA of MCT1 in SOD1 (G93A) or SOD1 (WT) spinal cord at different disease stage. Indeed the disregulation is post-transcriptional and the decrease in MCT1 is not only secondary to oligodendrocyte cell loss. Fig. 10

Image Unavailable
MCT1 expression in SOD1(WT), SOD1(G93A)
and SOD1(A4V) remained unaffected

Fig. 10
Image Unavailable
MCT1 m-RNA levels in SOD1 (G93A) mice
throughout disease progression.

Therefore the results suggest that loss and alterations of MCT1 in oligodendrocytes may contribute to motor neuron degeneration in ALS mice models and in ALS patients.
MCT1 in oligodendrocytes are necessary for the supply of lactate to neurons. Their disruption results in motor neuron dysfunction and degeneration [2]. When adding lactate to organotypic cultures, cell loss was prevented. In MCT1+/- mice, oligodendrocytes do not degenerate, this also was seen when spinal cord cultures were exposed to MCT1i or were glucose deprived. Therefore, MCT1 is necessary for the proper transport of energy to axons and their disruption lead to axonal dysfunction and neuron degeneration. It was found that MCT1 was reduced in motor cortex of patients with ALS and in the spinal cord, specially the ventral horn of SOD1 (G93A) mice. Suggesting that MCT1 is down regulated or maybe the death of oligodendrocytes are replaced by immature oligodendrocytes which do not have the transporter, therefore being unable to supply metabolites to neurons and leading to axonal damage and neuronal loss.

3. Selective removal of mutant SOD1

Oligodendrocytes are very delicate cells that are prone to damage due to the expression of genes linked to neurodegeneration [11]. Kang et al [3] selectively removed mutant human SOD1 (G37R) form NG2+ cells in Pdgfra-creER; loxSOD1 (G37R) mice [12], to determine whether oligodendrocytes contribute to the disease onset and progression. This resulted in mice which had a delayed disease onset and prolonged survival. Nevertheless the survival rate was not prolonged, because it was only due to the delayed disease onset. These mice also showed less increase in the number of astrocytes due to the destruction of neurons in the CNS, and less glial activation in the spinal cord. The removal of SOD1 (G37R) mutation decreased SOD1 (G37R) gene expression in NG2+ cell population by about 43%, but its expression was maintained in motor neurons. Therefore the new oligodendrocytes produced will come from the SOD1 (G37R) mutation deletion, so they will not show the mutation.
From Lee et al [2], we know that metabolic support to oligodendrocytes is through MCT1 transporter, and MCT1 expression is reduced in SOD1 (G93A) mice by end stage of the disease. Therefore the removal of SOD1 (G37A) from NG2+ cells preserved MCT1 expression in SOD1 (G37R) mutation deletion [((bibcite Kang).

1. Philips T, Bento-Abreu A, Nonneman A, Haeck W, Staats K, Geelen V, Hersmus N, Kusters B, Van Den Bosch L, Van Damme P, Richardson W, Robberecht W. Oligodendrocyte dysfunction in the pathogenesis of amyotrophic lateral sclerosis. Brain. (2013); 136: 471-482.
2. Lee Y, Morrison BM, Li Y, Lengacher S, Farah MH, Hoffman PN, Liu Y, Tsingalia A, Jin L, Zhang P, Pellerin L, Magistretti PJ, Rothstein JD. Oligodendroglia metabolically support axons and contribute neurodegeneration. Nature. (2012); 487: 433-448.
3. Kang SH, Li Y, Fukaya M, Lorenzini I, Cleveland DW, Ostrow LW, Rothstein JD, Bergles DE. Degeneration and impaired regeneration of gray matter oligodendrocytes in amyotrophic lateral sclerosis. Nature Neuroscience (2013); 16 (5): 571-579.
4. Kang SH, Fukaya M, Yang JK, Rothstein JD, Bergles DE. NG2+ CNS glial progenitors remain committed to the oligodendrocyte lineage in postnatal life and following neurodegeneration. Neuron. (2010); 68:668–681.
5. Zawadzka M, Rivers LE, Fancy SP, Zhao C, Tripathi R, Jamen F, et al.
CNS-resident glial progenitor/stem cells produce Schwann cells as well as oligodendrocytes during repair of CNS demyelination. Cell Stem Cell (2010); 6: 578–90.
6. Rinholm JE, Hamilton NB, Kessaris N, Richardson WD, Bergersen LH, Attwell D. Regulation of oligodendrocyte development and myelination by glucose and lactate. J Neurosci. (2011); 31:538–548.
7. Murray CM, Hutchinson R, Bantick JR, Belfield GP, Brazma D, Bundick RV, Cook ID, Craggs RI, Evans LR, Edwards S, Holness E, Jackson CG, Kingston LP, Sullivan M, Taylor-Fishwick DA, Walker PC, Wilkinson DJ, Wright A, Donald DK. Monocarboxylate transporter MCT1 is a target for immunosuppression. NatChem Biol. (2005); 1:371–376.
8. Edgar JM, McLaughlin M, Werner HB, McCulloch MC, Barrie JA, Brown A, Faichney AB, Snaidero N, Nave KA, Griffiths IR. Early ultrastructural defects of axons and axon-glia junctions in mice lacking expression of Cnp1. Glia. (2009); 57:1815–1824.
9. Kusaka H, Hirano A. Fine structure of anterior horns in patients without amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. (1985); 44:430–438.
10. Morrison BM, Shu IW, Wilcox AL, Gordon JW, Morrison JH. Early and selective pathology of light chain neurofilament in the spinal cord and sciatic nerve of G86R mutant superoxide dismutase transgenic mice. Exp Neurol. (2000); 165:207–220
11. Bu J, Akhtar N, Nishiyama A. Transient expression of the NG2 proteoglycan by a subpopulation of activated macrophages in an excitotoxic hippocampal lesion. Glia. (2001); 34:296–310.
12. Garbern JY, Yool DA, More GJ, Wilds IB, Faulk MW, Klungmann M, Nave KA, Sistermans EA, van der Knaap MS, Bird TD, Shy ME, Kamholz JA, Griffiths IR. Patients lacking the major CNS myelin protein, proteolipid protein 1, develop length-dependent axonal degeneration in the absence of demyelination and inflammation. Brain. (2002); 125:551–561.

Add a New Comment
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License