The molecular mechanisms mediating chronic neurodegeneration and MS progression are not well understood. OxPC are neurotoxic byproducts of oxidative stress and OxPC accumulation prominently associates with all stages of MS17, 20, 21, 26. Here, we report that OxPC deposition in the mouse SCWM induced a chronic focal lesion with features akin to CAL from P-MS including compartmentalized neuroinflammation, demyelination, and neurodegeneration, as well as endogenous OxPC accumulation. The severity of chronic neurodegeneration and endogenous OxPC accumulation in these CAL-like lesions were significantly increased by aging and by microglia depletion. An increase of IL-1β levels was associated with lesion chronicity and with conditions that exacerbated chronic neurodegeneration and endogenous OxPC accumulation, suggesting a link between OxPC mediated pathology and IL-1β activity. Indeed, E06+ OxPC immunoreactivity as well as the loss of NFH+ axons were significantly lower in CAS1/4−/− mice, which have deficiency in pro-IL-1β processing52, 53. Conversely, IL-1R1 blockade ameliorated axon loss and endogenous OxPC accumulation in chronic lesions. These results together with the observation that IL-1β and OxPC accumulated in CAL from P-MS brains, suggested that aberrant IL-1β signaling and OxPC contributes to chronic neurodegeneration in P-MS.
Although OxPC accumulation17–22 and inflammation associated oxidative injury8, 9, 14, 15, 18, 23–25 are major features of MS pathophysiology, their functional involvement in P-MS is not well understood because commonly studied experimental models of MS such as EAE or LPC induced demyelination have pathologies that involve relatively low levels of OxPC and oxidative injury18, 19. Here, we show that a one-time deposition of POVPC, a type of OxPC upregulated in MS lesions26, promoted chronic unremitting compartmentalized injury in the SCWM of mice. Chronic OxPC lesions had significant neurodegeneration in the form of axonal loss, persistent demyelination/remyelination failure, as well as accumulation of endogenous OxPC and inflammatory microglia/monoMac that upregulated NADPH oxidase and IL-1β. These pathological responses are similar to CAL disease activity in P-MS, which also involves axon loss56, ongoing demyelination57, microglia/monoMac reactivity9, and NADPH oxidase upregulation25. Moreover, we found significant accumulation of OxPC and IL-1β with reactive phagocytes in CAL from P-MS brains. These observations indicate that OxPC mediated chronic lesions recapitulated key aspects of P-MS pathophysiology. Notably, the exacerbation of OxPC mediated chronic neurodegeneration in middle-aged mice is evidence that aging associated defects in OxPC mitigation contributes to the acceleration of chronic neurodegeneration as people with P-MS become older. Thus, this model will be useful for uncovering new mechanisms that regulate chronic neurodegeneration in P-MS and aging.
The concentration of OxPC deposited in active lesions or CAL from MS have not been measured systematically and only one study estimated there are approximately 0.6 µg of OxPC per mg of protein isolated from MS brain homogenate58. In this study, chronic lesions developed after the deposition of 5 µg of POVPC, which is likely within the range of OxPC concentrations potentially found in MS lesion microenvironments. Using the OxPC specific and neutralizing E06 antibody29, 30, 38, we previously found minimal OxPC in the focal SCWM lesion at 1 day post POVPC injection26, which indicates that most if not all of the injected POVPC quickly reacts with the tissue. Thus, the detection of E06+ OxPC within focal OxPC lesions after 42 and 100 days reflects significant endogenous OxPC production, potentially due to similar mechanisms found in MS lesions. Notably, OxPC accumulation was minimal in chronic LPC or EAE lesions, highlighting that the pathology in these models may be different from that of chronic OxPC lesions and OxPC+ CAL from P-MS.
The cause of the endogenous OxPC accumulation in MS lesions remains relatively unknown. Reactive microglia/monoMac are major populations of immune cells found in CAL of P-MS and they express NADPH oxidase17–19, 25. Thus, they have the capacity to produce ROS, which can promote nonenzymatic oxidation of PC into OxPC30 in the lesion microenvironment. Similarly, microglia and monoMac with elevated NADPH oxidase expression were also the predominant immune cells in chronic OxPC lesions, suggesting they may contribute to lesion chronicity by producing ROS which leads to lipid peroxidation and endogenous OxPC formation, which in turn generate a feed-forward loop to generate more ROS and OxPC30. Alternatively, NADPH oxidase upregulation by microglia/monoMac in chronic lesions may not fully reflect in situ ROS/OxPC generation as PC oxidation can also be initiated enzymatically by 12/15-lipoxygenase59, 60, which is also upregulated in MS61. Both of these possibilities may be investigated by generating transgenic mice with microglia/monoMac deficient in NADPH oxidase62 or 12/15-lipoxygenase63 expression. Nevertheless, since IL-1β was significantly elevated in chronic OxPC lesions and in CAL from P-MS, it may be involved in sustaining microglia/monoMac reactivity, oxidative stress, and lipid peroxidation in the lesion microenvironment. In support of this hypothesis, we previously showed IL-1β deposition in the SCWM directly promotes acute neuroinflammation and endogenous OxPC formation26 whereas here we found blocking IL-1R1 signaling reduced OxPC mediated chronic neurodegeneration. These findings, together with recent evidence showing that OxPC can also enhance IL-1β production in both primed myeloid cells50, 51, 64 and mouse peritoneal macrophages29 suggest a potential positive feedback loop between OxPC and IL-1R1 mediated inflammatory responses. Thus, strategies to inhibit IL-1β signaling in P-MS may help to interrupt the vicious cycle of OxPC generation, neuroinflammation, and neurodegeneration. It is important to note that OxPL are abundantly generated whenever cells undergo apoptosis cell death, thus further contributing to this positive feed-forward destructive cycle65, 66.
CAL associate with chronic neurodegeneration and disability progression in people with P-MS8, 57, 67. They are classified by a relatively inactive lesion core surrounded by a slowly-expanding lesion rim with ongoing myelin loss and accumulation of lipid-laden microglia/monoMac1, 41, 67. The relative composition of microglia versus monoMac in CAL remains unclear and is difficult to determine as the expression of microglia specific markers such as TMEM119 or P2RY12 becomes less stable in reactive microglia68. While microglia and monoMac in CAL are largely characterized as proinflammatory cells9, 15, 68, their functional contribution to chronic neurodegeneration remains uncertain. By fate mapping tdTomato+ microglia in chronic OxPC lesions, we found that unlike in acute lesions26, 27, 42, tdTomato− monoMac re-emerged as a significant immune cell population in chronic OxPC lesion. Importantly, the amount of monoMac versus microglia was significantly increased within chronic lesions of middle-aged mice compared to lesions from young mice, which is consistent with recent findings showing that chronic injury and aging drive the accumulation of disease associated monoMac in the CNS47, 69. Importantly, we found that middle-aged mice with greater monoMac accumulation and mice with microglia depletion had greater OxPC and IL-1β accumulation as well as worse chronic neurodegeneration compared to young mice and mice without microglia depletion, respectively. These findings together with studies showing that microglia mediated phagocytic processing of harmful lipids becomes dysregulated by aging26, 27, 70–73, implicate that microglial loss and/or their dysregulation caused by chronic disease activity and aging can facilitate the acceleration of chronic neurodegeneration in P-MS.
In summary, we report that OxPC deposition in the CNS induces pathology similar to CAL from P-MS including endogenous lipid peroxidation, unremitting neuroinflammation, and chronic neurodegeneration. Aging and microglial dysregulation in part through inflammatory IL-1β signaling exacerbated chronic neurodegeneration. Promoting OxPC neutralization and clearance, as for example by targeting with the E06 antibody, and/or promoting homeostatic microglia repopulation may be effective approaches to slow and/or halt chronic neurodegeneration in P-MS.
ONLINE METHODS
MS specimens
Post-mortem human brain tissues were obtained from seven patients diagnosed with clinical and neuropathological P-MS according to the revised 2010 McDonald’s criteria67, 74, with full ethical approval (BH07.001, Nagano 20.332 - YP) and informed consent as approved by the CRCHUM and University of Montreal research ethics committee. As previously described75, 76, autopsy samples were cryopreserved, and lesions were classified using Luxol fast blue and H&E staining.
Mice
All experiments were conducted with ethics approval (protocol number 20220103) from the Animal Care Committee at the University of Saskatchewan under regulations of the Canadian Council of Animal Care. Female 6wk and 52wk old C57Bl/6J mice were acquired from Jackson Laboratories for in vivo experiments. CX3CR1creER (strain 021160) mice, Ai9TdTom mice (strain 007909), and Rosa26iDTR mice (strain 007900) from The Jackson Laboratory were bred in the Lab Animal Services Unit at the University of Saskatchewan to produce male and female CX3CR1CreER:Ai9tdTom mice and CX3CR1CreER:Rosa26iDTR mice for microglial fate mapping and depletion studies. Female 6-10wk old C57BL/6NJ (CAS1/4+/+, strain 005304) and B6N.129S2-Casp1tm1Flv/J (CAS1/4/− strain 016621) mice from Jackson Laboratories were used for IL-1β studies. Mice were maintained on a regular diet in low humidity environment on a 12-hr light/dark cycle at 21 to 23 degrees Celsius with unlimited access to food and water. Mice and littermates were randomly assigned to different experimental groups.
Fate mapping microglia/monoMac
Six-week-old and 52-week-old CX3CR1CreER:Ai9tdTom mice were intraperitoneally injected with 2 mg of tamoxifen (20 mg/ml; T5648, Sigma) dissolved in corn oil (C8267) once a day for 3 consecutive days to induce tdTomato expression in all CX3CR1+ mononuclear phagocytes. Mice were then used 4 weeks after tamoxifen injection which is when tdTomato expression only labels long lived CNS resident microglia and macrophages but not monoMac derived from infiltrating monocytes.
Microglia depletion
Newly weaned CX3CR1CreER:Rosa26iDTR mice were injected intraperitoneally with 2 mg tamoxifen (20 mg/ml; T5648, Sigma) dissolved in corn oil (C8267) once a day for 5 consecutive days to induce DTR expression on microglia and were used for experiments 3 weeks after tamoxifen injection. For microglia depletion during acute lesion phase, tamoxifen treated CX3CR1CreER:Rosa26iDTR mice were injected intraperitoneally with PBS or 1 µg of DT every other day from days 0 to 7 after OxPC deposition in the SCWM. For microglia depletion during chronic lesion phase, tamoxifen treated CX3CR1CreER:Rosa26iDTR mice were injected intraperitoneally with PBS or 1 µg of DT every other day from days 35 to 42 after OxPC deposition.
Anakinra blockade of IL-1R1
Six-week-old C57Bl/6J mice were intraperitoneally injected daily with 100 µl of PBS or PBS containing 1 mg/kg anakinra (HY-108841, MedChemExpress) from days 28–42 following OxPC deposition in the SCWM.
Spinal cord surgery
The surgical procedure for OxPC spinal cord injection was performed as described from previous studies26, 27, 77. Briefly, mice were anesthetized with ketamine and xylazine and injected stereotactically with 0.5 µl PBS containing 10 mg/ml POVPC (Avanti Polar Lipids, 870606P) into the ventrolateral SCWM between the T3 and T4 vertebra. After the injection, the needle was left in place for 2 min to prevent back flow, and then the mouse was sutured and placed in a thermally controlled environment for recovery. Alternatively, equivalent amount of LPC (Sigma, L1381) was injected in the SCMW using the same approach.
EAE and tissue isolation
Eight- to 10-week-old female C57Bl/6J mice were subcutaneously inoculated with 50 µg of MOG 35–55 peptide (Protein and Nucleic Acid Facility, Stanford University School of Medicine) in 100 µl of complete Freund’s adjuvant supplemented with 4 mg/ml heat-inactivated Mycobacterium tuberculosis H37Ra (Sigma-Aldrich), in which 50 µl emulsion was deposited on each side of the tail base. Intraperitoneal injection of pertussis toxin (300 ng per 200 µl; 180, List Biological Laboratories) was performed days 0 and 2 after MOG immunization. Mice were monitored and scored daily on a scale of 0–15. EAE mice during chronic disease (day 45) were euthanized with intraperitoneal injections of ketamine and xylazine. Fifteen ml of PBS was then perfused via cardiac puncture, and the cerebellum and spinal cord were collected. The cerebellum was frozen in Optimal cutting temperature polymer (Leica), whereas the spinal cord was processed as stated above for the spinal cord injections. Cerebellar sections (sagittal) and the spinal cord sections (longitudinal) were cut into 20-µm sections using a cryostat and collected onto Superfrost Plus microscope slides (VWR) and stored at − 20°C before analysis.
Spinal cord tissue isolation for histology and microscopy analysis
Mice were euthanized with intraperitoneal ketamine and xylazine overdose after 7-, 42-, or 100-days post-surgery. Ten ml of PBS followed by 10 ml of 4% paraformaldehyde in PBS were perfused via cardiac puncture. The spinal cord was then dissected from the back of the mouse, and the tissue containing the T3-T4 inject site was collected into 4% paraformaldehyde in PBS for fixation overnight at 4 degrees. Thereafter, spinal cords were transferred to 30% sucrose solution for dehydration for at least 48h and frozen in FSC 22 Frozen Section Media (Leica). With a cryostat (ThermoFisher Scientific), spinal cord tissue was cut into 20 µm coronal sections and collected on to Superfrost Plus microscope slides (VWR). Tissues were stored at -20 degrees prior to staining and analysis.
Antibodies
The following primary antibodies were used for immunofluorescence microscopy: mouse IgM E06 anti-OxPC (5 µg/ml, generously provided by Witztum and Tsimikas labs), rabbit anti-human/mouse IBA1 (1:1000, Wako 019-19741), chicken anti-human/mouse IBA1 (1:1000, Synaptic Systems 234 009), rat anti-mouse MBP (1:200, Abcam ab7349), rat anti-mouse CD11b (1:200, ThermoFisther 14-0112-82), goat anti-human/mouse OLIG2 (1:200, R&D Systems AF2418), rabbit anti-mouse NFH (1:1000, Encor Biotechnology RPCA-NF-H), chicken anti-mouse NFH (1:1000, Encor Biotechnology CPCA-NF-H), rat anti-mouse iNOS (1:100, ThermoFisher 14-5920-82), mouse anti-human/mouse IL-1β (1:100, Cell Signaling 12242S), rat anti-human/mouse CD45 (1:200, ThermoFisher MA5-17687), rabbit anti-mouse cleaved IL-1β (1:100 Cell Signaling 63124S), Rabbit anti-mouse Arg1 (1:200, Cell Signaling 93668S, rat anti-mouse CD16/32 (1:100, BD Pharmingen 553141). rabbit anti-mouse P22phox (1:200 Cell Signaling 37570S), chicken anti-human/mouse GFAP (1:1000, Biolegend 829401), rabbit anti-human/mouse βAPP (1:200, Thermo Fisher Scientific 36-6900), rabbit anti-human/mouse Ki67 (1:200, Abcam ab15580), and rabbit anti-mouse p16INK4A (1:100, Cell Signaling 29271S).
The following secondary antibodies from Jackson ImmunoResearch were used at 1:400 dilution: Alexa Fluor 488 donkey anti-mouse IgM, Alexa Fluor 488 donkey anti-mouse IgG, Cyanine Cy3 donkey anti-chicken IgY, Alexa Fluor 647 donkey anti-rat IgG, Cyanine Cy3 donkey anti-rat IgG, Alexa Fluor 488 donkey anti-goat IgG, Alexa Fluor 647 donkey anti-rabbit IgG.
Mouse spinal cord histology
For eriochrome cyanine (EC) and neutral red (NR) visualization of serial spinal cord lesions, spinal cord sections were stained as previously described26–28. Alternatively, serial spinal cord lesions were also visualized by immunofluorescence labeling of MBP. Brightfield images were then acquired using the Olympus VS110 Slidescanner with a 10x 0.4 NA air objective. Lesion ROIs were drawn based on demyelinated areas in the SCWM that have lower EC staining and higher NR staining and their total areas were quantified using the CEllSens Dimension software (Olympus). Lesion volume was estimated by multiplying the distance separating each serial section (400 µm) by the sum of lesion areas in the serial spinal cord sections from each sample.
Labeling tissues for immunofluorescence confocal microscopy
As previously described26–28, slides with mouse spinal cord samples were warmed to room temperature (RT) for 10 min. When MBP staining is required, slides were delipidated by successive wash of 50%, 70%, 90%, 95%, 100%, 95%, 90%, 70%, and 50% ethanol. Then, samples were rehydrated in PBS for 10 min and permeabilized with 0.2% Triton-X100 in PBS for 10 min. Samples were then blocked with donkey blocking solution (PBS, 10% donkey serum, 1% BSA, 0.1% cold fish stain gelation, 0.1% Triton X-100, 0.05% Tween-20) for 1h at RT or overnight at 4 degrees. Alternatively, E06 antibody staining, 5 µg/ml of purified Rat Anti-Mouse CD16/CD32 Fc blocking antibody (5 µg/ml, BD Pharmingen) was added to the blocking buffer. After blocking, samples were incubated with primary antibodies in antibody dilution buffer (PBS, 1% BSA, 0.1% cold fish stain gelation, 0.1% Triton X-100) for overnight incubation at 4 degrees. Samples were then washed 3 times, 5 min each with PBS and 0.2% Tween-20 and incubated with secondary antibodies and 1 µg/ml of DAPI resuspended in the antibody dilution buffer for 1h at RT. For samples with high potential of autofluorescence, slides were also blocked using the TrueBlack Lipofuscin Autofluorescence Quencher (Biotium) in accordance with manufacturer’s instructions. Finally, slides were washed 3 times using PBS with 0.2% Tween-20, 5 min each, and coverslips were mounted onto the slides using Fluoromount-G solution (SouthernBiotech).
For post-mortem MS tissue samples, after slides were warmed to RT, they were fixed with 4% paraformaldehyde for 10 min, then washed in PBS for 10 min to remove excess PFA. The remaining steps were the same as the mouse spinal cord samples.
Immunofluorescence microscopy
Immunofluorescence images were acquired using the Leica TCS Sp8 laser confocal microscope at RT, using the 10x 0.40 NA air object or the 25x 0.5 NA water objective. The 405 nm, 488 nm, 552 nm, and 640 nm lasers were used to excite the fluorophores from antibodies bound to samples and detected by two low dark current Hamamatsu PMT detectors and two high sensitivity hybrid detectors. Images were acquired in 8-bits, in a z-stack using unidirectional scanning, 1 airy unit pinhole, 0.75x zoom, and 0.57 µm optical sections and 2048 x 2048 pixels xy resolution. Alternatively, images were acquired using the Zeiss LSM700 confocal microscope with a 20x 0.8 NA air objective using similar settings. Images were also acquired using the Zeiss Axio Observer 7 widefield microscope with Colibri 7 LED illumination and a 25x 0.85 NA water objective, followed by constrained iterative deconvolution processing. Equal laser, gain, offset, and exposure settings to maximize contrast and minimize saturation were consistently used for all samples within experiment sets. A sample slide stained with only the secondary antibodies and DAPI was used for each experiment to control for non-specific secondary immunofluorescence. Leica Application Suite X or Zeiss Zen Black software was used for image acquisition, ImageJ was used for image threshold and particle analysis.
Image analysis
Z-stack confocal images of spinal cords were analyzed with ImageJ (Fiji, NIH) as previously described26–28. Briefly, maximum intensity projections were created for each channel/marker z-stack and converted from 8-bit to RGB. The lesion ROI or equivalent area in the contralateral normal appear white matter (NAWM) was drawn while the area outside the ROI was not analyzed. Positive signal was determined using the color brightness threshold set consistently using a predetermined value by comparing the secondary antibody-stained control and NAWM. Lesion ROIs were drawn based on markers that define the SCWM lesion area such as CD16/32, IBA1, or MBP. The analyze particles function was then used to quantify the positive signals in each ROI. To avoid bias, the same threshold values for setting the positive signal, as well as the size and circularity settings for particle analysis were used for all samples in each experimental set. For representative images shown, maximum intensity projection of each channel/marker in a z-stack were merged and displayed using pseudo colors. Only brightness and contrast settings were adjusted in ImageJ, and consistently between samples for better displaying the images.
Statistics and reproducibility
Data were collated in Microsoft Excel and graphs were generated using GraphPad Prism 10.1 (LaJolla, CA). Data shown are the individual data points where each point on a graph represents a separate mouse. The mean ± SD are also shown. No sample size calculation was performed. Sample size was determined based on previously published results26–28 and based the cost of experiment, feasibility of the experiment, as well as the availability of sex and aged matched mice. No data was excluded from the analyses. Sample sizes are reported in the figure legends and only one measurement is recorded per sample. Littermate mice were randomly selected for each experimental condition and treatment. Blinding was not conducted. For analysis of statistical significance between the means of two or more treatment groups against the control group, one-way ANOVA with Tukey’s multiple comparison test was used. Two-tailed, unpaired t-test was used to compared data with only two groups. Kolmogorov-Smirnov test was used to verify the normal distribution of data. Specific P-values are reported in each figure where results are considered statistically significant where p < 0.05.