Corticohippocampal Dysfunction In The OBiden Mouse Model Of Primary Oligodendrogliopathy

Despite concerted efforts over decades, the etiology of multiple sclerosis (MS) remains unclear. Autoimmunity, environmental-challenges, molecular mimicry and viral hypotheses have proven equivocal because early-stage disease is typically presymptomatic. Indeed, most animal models of MS also lack defined etiologies. We have developed a novel adult-onset oligodendrogliopathy using a delineated metabolic stress etiology in myelinating cells, and our central question is, “how much of the pathobiology of MS can be recapitulated in this model?” The analyses described herein demonstrate that innate immune activation, glial scarring, cortical and hippocampal damage with accompanying electrophysiological, behavioral and memory deficits naturally emerge from disease progression. Molecular analyses reveal neurofilament changes in normal-appearing gray matter that parallel those in cortical samples from MS patients with progressive disease. Finally, axon initial segments of deep layer pyramidal neurons are perturbed in entorhinal/frontal cortex and hippocampus from OBiden mice, and computational modeling provides insight into vulnerabilities of action potential generation during demyelination and early remyelination. We integrate these findings into a working model of corticohippocampal circuit dysfunction to predict how myelin damage might eventually lead to cognitive decline.

Win-shift T-maze foraging: companion to novel object recognition 4 . Spontaneous alteration behavior of mice is used to identify deficits in working memory and hippocampal dysfunction. T-shaped arena 30cm long x 10cm wide x 20cm high, with a 7cm central partition between the top arms, and guillotine doors that can be lowered to isolate these arms ( Supplementary Fig. S6B). Mice are placed at the base of the T facing away from the top arms. The mice turn and traverse the stem, enter one of the top arms, and the associated door is lowered to allow 30s exploration of the arm. The mice are returned to their home cage for 2min, and the trial is repeated. Subsequent trials are repeated 10 times/day for 3 days. The maze is cleaned with 70% ethanol between trials.
Tail suspension: mice are suspended vertically by taping their tail to a standard retort stand clamp elevated 30cm above a table and video recorded 5 . Plastic tubing is placed over the tail to stop mice crawling up their tail during testing. Mice hang suspended for 6min. Mice are returned to their home cage. Total time spent moving versus time immobile (not moving other than that needed for adequate respiration) is subsequently measured from the video using a stopwatch and is interpreted as "giving up" which is a proxy for depression-like behavior.
Forced swim: two day (acclimation and test days) companion to tail suspension 6 . Each day, mice are placed in a 12cm diameter cylinder containing a 30cm column of water at 25°C for 10min on the first day and 6min on the second day. Mice are video recorded to determine time spent immobile versus swimming/struggling.
Barnes Maze: opaque white circular platform, 91cm diameter, with 30cm x 5cm holes around the circumference (7.6cm from the edge) 7 . An escape/goal box measuring 10cm x 10cm x 5cm is placed under one of the holes ( Supplementary  Fig. S6D). Mice associate the goal box with escape by placement inside the goal hole. Testing lasts 11 days, with 3 x 5min trials/day/mouse as follows: days 1-5, training to find and enter the goal hole. Day 6, a 90° clockwise maze rotation probe day. Mice should return to the training day location. Day 7, retraining day with the maze in the original orientation. Day 8, curtain probe day to block visual cues that the mice use to find the goal box, which determines if intrinsic property cues on the maze allow the mice to locate the goal hole (these cues must be eliminated). Day 9, mice are retrained as for Days 1-5 and 7. Day 10, the goal hole is rotated 180° from the original location. Mice are trained to the new location to determine how easily they relearn. Day 11, probe day to determine if the mice remember the day 10 goal hole location. The latency to reach the goal hole and the number of errors (defined as attempting to enter an incorrect hole) are recorded. Trials are video recorded for analysis using Ethovision tracking software (Noldus).
Contextual/cued fear conditioning: cued and contextual fear conditioning, fear extinction and reinstatement/renewal testing 8 . Conditioning chambers include contextually unique environment, removable shock grid floor, house light, speaker, and cue light (Coulbourn Instruments, Whitehall, PA). Prior to experimentation foot shock is optimized for each mouse. The intensity (0.3-1mA) and duration (0.5-2s) are varied to find the minimum shock needed to elicit an avoidance response. Contextual based conditioning is based on shock presentation after mice have habituated to the chamber and cue based conditioning pairs auditory or visual stimuli with foot shock.
Contextual fear conditioning is a 2 day procedure where animals are placed in a contextually unique chamber (normal lighting conditions without food or water during testing). After 3min acclimation, mice receive 3-5 foot shocks (0.5-2s) with varying interstimulus intervals (30-120s). Mice are returned to their home cages. After 24h, mice are returned to the chamber and video recorded for 10min to capture freezing behavior (without additional foot shocks).
Cued fear conditioning is a 2 day procedure of acclimation and exposure to 5-15 trials of cue-shock presentation (on day 1). Cues are short duration lights or tones (85dB, 16kHz, 20s) followed by foot shock (0.3-1mA) in the final 0.5-2s of cue presentation. After 24h, mice are tested for 3-10min for freezing behavior (no foot shock) to either the context of the chamber or presentation of a cue for 3-10min in a contextually distinct chamber.
Fear extinction testing is the same procedure for cued and contextual conditioning, and involves daily extinction sessions (no foot shock) for 12 days that are identical to the day 2 session for each test.
Reinstatement/renewal testing involves day 1 of cued fear conditioning in a distinct context 'A' in the chamber. Day 2, mice are exposed to two conditions, 3h apart: 1) context 'B' (distinct from context 'A') where mice are presented with 15 trials of light/tone (no foot shocks); 2) context 'A' for 30min (no foot shocks). Two days after conditioning (day 4), mice are separated into 2 groups and exposed to either context 'A' or 'B' (no foot shocks) or 15 trials of light/tone (no foot Supplementary Information Corticohippocampal Dysfunction In The OBiden Model Of Primary Oligodendrogliopathy Daniel Z. Radecki, Elizabeth L. Johnson, Ashley K. Brown, Nicholas T. Meshkin, Shane A. Perrine and Alexander Gow shocks). Freezing behavior is measured to determine cued extinction retention and fear renewal. Mice are otherwise housed in their home cage.
To stain axons throughout the CNS we used the following reagents for Bielschowsky modified silver stain: 10% silver nitrate 9 , 10% silver nitrate with ammonium hydroxide (added drop wise with stirring until the silver oxidizes then turns clear), developer stock solution -80ml distilled H2O, 20ml formaldehyde, 0.5g citric acid, 2drops 14N nitric acid, working developer solution -16drops developer stock, 16drops ammonium hydroxide and 100ml distilled water, 5% (w/v) sodium thiosulfate, and ammonium water made of 16 drops ammonium hydroxide in 100ml water. All glassware was acid washed before staining. For staining, the 10% silver and 10% silver + ammonium were heated, covered, to 40°C and the slides thawed in 1X PBS followed by dehydration through ethanol series to xylene and back to distilled water. Slides were then moved to 10% silver for 1min at 40°C, washed in distilled water and placed in 10% silver + ammonium for 10min at 40°C. Next, slides were moved to the working developer solution where fiber tracks turn black during differentiation. To stop differentiation, slides were washed 3 x 2min in ammonium water, then washed in 5% sodium thiosulfate for 5min to chelate residual heavy metals. Nuclei were counterstained with Hematoxylin.
A pilot experiment in the brain from a single double heterozygote male mouse, McreG::tdTomato (B6;129S6-Gt(ROSA)26Sor tm9(CAG-tdTomato)Hze /J), was used to determine the proportion of mature perivascular oligodendrocytes that could absorb sufficient tamoxifen from 2 successive daily gavages to activate cre-mediated recombination of the tdTomato transgene. Vibratome sections (100µm) were permeabilized with 1% Triton X-100 and labeled with mouse anti-CC-1 antibodies in TBSGBA (1:100 ,EMD Millipore), and through-focus images were generated at 40X magnification from confocal image stacks for analysis in imageJ. The number of EGFP + /CC1 + oligodendrocytes expressed as a proportion of total CC1 + cells was expressed as a function of the distance from blood vessels (center-to-center measurements).

Entorhinal cortex AIS image analysis
Ten micrometer image stacks of 25-30 images (0.3µm spacing) were captured using a Leica DM5500 microscope (Leica Microsystems Inc, Buffalo Grove, IL) equipped with a Plan Apo 40x/0.85 coverslip correction lens, Yokagawa spinning disk confocal with Melles-Girot argon ion laser (McBain Systems, Westlake Village, CA), Orca R2 digital camera (Hamamatsu Corp, Bridgewater, NJ) and Metamorph software (version 7.6, Lake Zurich, IL). Maximum intensity projection for extended focus RGB images of ankyrin-G / NeuN / DAPI triple-positive cortex were analyzed in Fiji/ImageJ 11 . Each ankyrin-G + AIS in the analysis conformed to two criteria: the proximal end was juxtaposed to a NeuN/DAPI double-positive cell body; the distal end had a tapering railroad track appearance and did not terminate with a truncated square end. The freehand line tool was used to measure AIS length, which accommodates in-plane nonlinearity.
At least five AIS profiles were measured/image (213µm x 163µm fields), with six images of left and right cortices from three matched 10µm cryostat sections/mouse. The AIS lengths were plotted for each mouse as a cumulative frequency distribution (0-40µm, 2µm bins) and curves were averaged by genotype and fit with a Gaussian curve for statistical analysis.

Hippocampus AIS analysis
Image stacks of the CA1 layer in dorsal (DHC) and ventral (VHC) hippocampus were captured as for cortex, except they comprised the central 8µm of image stacks from 10µm cryostat sections. The CA1 layer is defined as the cell dense layer directly dorsal (DHC) or lateral (VHC) to mid-dentate gyrus and representative of overall CA1 appearance to avoid confusion with the CA2 region. Individual AIS profiles are difficult to identify and measure in the granular CA1 layer; thus, ≥10 analysis rectangles 700 (110µm) x 50 (one cell body width) pixels centered over CA1 were measured and averaged to generate fluorescence intensity plots for each color. These RGB plots were generated separately for left and right hippocampal slices from three cryostat sections and averaged/mouse for integration under the curves and the final intensities plotted as functions of distance normal to the CA1 layer. Cognate plots were averaged from three mice/genotype.

Western Blotting
Mice are decapitated and the brain rapidly removed, cut into 2mm coronal slices using a brain matrix and frozen on dry ice. For tissue punches, a 1.5mm disposable biopsy punch (Integra Miltex, Integra, York, PA) at room temperature is attached to tubing with filter paper and air pressure is used to expel tissue from the punch. Brain slices were placed on an inverted glass petri dish covered with filter paper wetted with 0.9% saline (sodium chloride, S640, Fisher). Brain regions were quickly punched and expelled into -80°C Eppendorf tubes kept closed on dry ice.
Samples were diluted in Phos-pres to 2X final concentration, then mixed 1:1 in general sample buffer [1g SDS (BP166, Fisher), 0.5g sucrose (BP220, Fisher), 0.98g Tris-HCL (BP153, Fisher), 25ml distilled H2O, pH6.8 with bromophenol blue to equal concentrations for each sample. Samples were vortexed 3 x 5s each, then heated at 60°C for 5min and then spun down for 30s. Eight, 10 or 12%, 1.5mm thick, SDS-PAGE gels were cast with 10 or 15 well combs and samples electrophoresed at 25mA/gel with water circulated cooling. PVDF membranes were equilibrated in transfer buffer containing 20% methanol for at least 1h prior to transfer. Gels were stacked onto membranes and transferred, with cooling, at 500mA for 1h.

Magnetic resonance and diffusion tensor imaging
For T2 structural scans, all MR images were acquired on a Bruker 7T Clinscan horizontal bore magnet with Siemens Syngo MR B15 software and a dual coil mouse brain surface coil. Mice were anesthetized using isoflurane vapor at 1.5-3.0% in medical air to maintain 70-90breaths/min, and the MRI holder was maintained at 32-34°C using a heated water circulation system. The T2 scan parameters were as follows: TE=61ms, TR=1,200ms, TA~42min (depending on gating), 16mm x 16mm FOV, 180° flip angle, Turbo factor=22, 2 averages and 3D acquisition for in-plane resolution of 125µm x 125µm. Images were exported to OsiriX software (version 8, Pixmeo SARL, Bernex, Switzerland) for analysis and reconstruction. The DTI images were acquired as follows: TE=30ms, TR=2,500ms, TA~30min (depending on gating), across 6 diffusion directions, 5 x 500µm thick slices with 1.2mm spacing and 2D acquisition. Two sets of scans were obtained, offset by 500µm to give greater coverage of the brain. Scans were imported into OsiriX and from the unprocessed 4D output, Eigen vector maps were calculated for each vector 1(l1), 2(l2) and 3(l3) and the FA was calculated as: For each region analyzed, 8-10 pixels were used to calculate an average FA across the pixels and then regions were averaged across slices to get an overall FA for each region and each animal.

Statistical Analysis
All statistical analysis were performed using Prism (version 7; Graphpad Software Inc., La Jolla, CA). For longitudinal behavioral testing, 2-way ANOVAs were used to compare genotypes across ages to identify changes within ages and within groups across ages if necessary. Western blot analysis used 2-way ANOVA again to identify changes between genotypes, but also to identify any changes between proteins within groups, especially with AIS proteins as changes in opposing directions could signify degenerative changes. Kaplan-Meier plots with X 2 analyses were used to compare median lifespan of mutant mouse strains. Data are reported as mean ± SEM unless otherwise noted.

NEURON simulations
Computational modeling was performed using the NEURON simulation environment, version 7.4 14 . The model is constructed with six identical dendrites connected individually to the soma. Each dendrite is comprised of proximal medial and distal segments. The soma is connected via a hillock segment to an axon with six AIS segments (three proximal and three distal) followed by an unmyelinated axon segment, the myelinated axon with 60 internodes and finally a small terminal segment. The internodes are taken directly from a previous model (ModelDB accession 122442) used to determine the function of myelin tight junctions 15,16 . Passive properties and active mechanisms are adapted from published sources, and the current study, and are listed in Supplementary Tables S2-4. Potassium channel densities in distal dendrites were adjusted to ensure approximately 50% reduction in amplitude 500µm from the soma 17 . Membrane resistance at the soma, Rm=21MW, was determined from the maximum slope of an I-V plot.
Simulations were performed at 37°C with extracellular Na + =160mM, K + =5mM and intracellular Na + =30mM, K + =140mM. We used d_lambda=0.01 for all compartments and ran simulations with fixed time steps=0.001ms and absolute tolerance=0.0001. All simulations began with 1ms initialization, followed by 7ms IClamp current injections applied at the midpoint of the soma. Threshold currents were determined iteratively to define the lowest current (±1pA) to generate a bAP and/or AP. The bAP amplitudes and voltage-time functions were measured in dendrite 6,389µm from the soma, and their magnitudes were not affected by lengthening the distal dendrite by 100µm (i.e. no segment boundary artifacts). The AIS voltage-time functions were measured at the midpoint of segments 1 (proximal) and 6 (distal) and the APs were measured at the midpoint of node 17. We also applied linear voltage ramps using VClamp (10-200V/s), with lengths defined by the bAP/AP threshold potentials. This method generated stimulus artifacts at ramp termination, but yielded qualitatively similar results to IClamp.

Neuropathology summaries of patients in the current study by Dr Roscoe Atkinson, MD
MS-4659: Neuropathological diagnosis: 1. Multiple sclerosis 2. Chronic/active multiple sclerosis plaque formation Date Completed: 12-15-07 Gross Description: High quality, high resolution colored digital images of both hemispheres and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The coronal sections show irregular demyelinating periventricular plaque formation throughout the extent of the body of both lateral ventricles. There are also smaller focal areas of demyelination in the left temporal white matter, left thalamus, basilar pons, and pontine tegmentum. There is no other apparent softening, discoloration, hemorrhage, mass, or other lesion. There is no atrophy of the cerebellar folia. Significant atherosclerosis is not present Microscopic description: A section of periventricular white matter is examined with the H&E stain and the luxol fast blue (LFB) myelin stain (8). A section of hippocampus and temporal lobe is also examined with the H&E stain (10). The sections show plaque formation with up to 80% axonal loss and 100% demyelination (LFB). There is near complete loss of oligodendrocytes. Associated gliosis is prominent in the bordering white matter. There are scattered macrophages with intracytoplasmic myelin seen along the edge of the plaque in the LFB-stained section. There is no evidence of perivascular lymphocytic cuffing. Comment: These findings are those of chronic/active multiple sclerosis (MS) plaque formation and are consistent with the diagnosis of multiple sclerosis. 1. Multiple sclerosis 2. Chronic multiple sclerosis plaque formation Date Completed: 12-14-07 Gross Description: High quality, high resolution colored digital images of both hemispheres and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The coronal sections show extensive irregular demyelinating periventricular plaque formation throughout the extent of the bodies of both lateral ventricles, the basal ganglia, brainstem, and cerebellar white matter. There is no other apparent softening, discoloration, hemorrhage, mass, or other lesion. The striatum, lentiform nuclei, and thalami are normal. There is no atrophy of the cerebellar folia. Significant atherosclerosis is not present. The spinal cord is also submitted and is grossly normal. Microscopic description: A section of periventricular white matter is examined with the H&E stain and the luxol fast blue (LFB) myelin stain (8). The hippocampus and temporal lobe are also examined with the H&E stain (13). The sections show demyelinating plaque formation with up to 80% axonal loss and 100% demyelination (LFB). There is a marked decrease in oligodendrocyte density. Associated gliosis is minimal. There is no evidence of macrophage activity or perivascular lymphocytic cuffing. Comment: These findings are those of chronic multiple sclerosis (MS) plaque formation and are consistent with the diagnosis of multiple sclerosis: MS-4725: Neuropathological diagnosis: 1. Multiple sclerosis 2. Chronic/active multiple sclerosis plaque formation Date Completed: 5-4-08 Gross Description: High quality, high resolution colored digital images of both hemispheres and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The coronal sections show diffuse patchy irregular demyelinating periventricular plaque formation throughout the extent of both lateral ventricles. There is also involvement of both lentiform nuclei. There is also a vertically oriented centralized area of demyelination in the basilar pons. There is no other apparent softening, discoloration, hemorrhage, mass, or lesion. The striatum, lentiform nuclei, and thalami are normal. There is no atrophy of the cerebellar folia. Significant atherosclerosis is not present. The dura and spinal cord are also present and are grossly normal. Microscopic description: A section of periventricular white matter is examined with the H&E stain and the luxol fast blue (LFB) myelin stain (10) The hippocampus (10) and temporal lobe (10) are examined with the H&E stain. The sections show plaque formation with up to 90% axonal loss and I 00% demyelination (LFB). There is a prominent decrease in oligodendrocyte density and extensive associated gliosis. There are scattered examples of associated perivascular lymphocytic cuffing. Macrophage activity is present but sparse. The neocortex contains normal neuronal cellularity and normal subcortical white matter. The hippocampus is also normal without evidence of neurodegeneration or hypoxia.

MS-4951: Neuropathological diagnosis:
1. Multiple sclerosis 2. Active multiple sclerosis plaque formation Date Completed: 1-7-12 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. There is no cerebral atrophy, cerebral edema, or herniation of the unci or cerebellar tonsils. The coronal sections show variable amount of irregular demyelinating periventricular plaque formations throughout the anterior body of the left lateral ventricle. There is similar plaque formation (0.6 x 0.4cm) involving right parietal lobe. There is no other apparent softening, discoloration, hemorrhage, mass, or lesion. The striatum, lentiform nuclei, and thalami are normal. There is no atrophy of the cerebellar folia. The basilar cerebral vasculature shows no significant atherosclerosis. Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The coronal sections show extensive irregular demyelinating periventricular plaque formation throughout most of the bodies of both lateral ventricles with more anterior extension in the right hemisphere. There is similar subcortical plaque formation (0.6 x 0.4cm) present in the left parietal lobe. There is no other apparent softening, discoloration, hemorrhage, mass, or lesion. The striatum, lentiform nuclei, and thalami are normal. There is no atrophy of the cerebellar folia. Significant atherosclerosis is not present.
Microscopic description: Sections of right periventricular white matter (8) are examined with the H&E stain and the luxol fast blue (LFB) myelin stain. Sections of hippocampus (6A) and temporal lobe (68) are examined with the H&E stain. Sections of periventricular white matter show plaque formation with 80% axonal loss, 80% oligodendrocyte loss, 80% demyelination (LFB) and prominent gliosis. There is evidence of macrophage activity and moderate perivascular lymphocytic cuffing. The hippocampus is normal without evidence of neurodegeneration or hypoxia. Comment: These findings are those of chronic/active multiple sclerosis (MS) plaque formation and chronic multiple sclerosis plaque formation.

No evidence of metastatic carcinoma.
Date Completed: 6-12-03 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined m a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The frontal, parietal, and temporal lobes show no atrophy. There is mild cerebral edema but no associated herniation of the hippocampal unci or cerebellar tonsils. There is no evidence of mass, lesion, hemorrhage, infarction, or metastatic disease. The substantia nigra shows normal pigmentation. The basilar cerebral vasculature shows no atherosclerosis. Microscopic description: The hippocampus with adjacent temporal lobe (11) is examined with the H&E stain. There is normal neuronal cellularity with no evidence of neuropathology, hypoxia, or metastatic disease. Comment: These findings are unremarkable and show no evidence of pathology.
HC-3540: Neuropathological diagnosis: 1. Essentially normal brain. Date Completed: 6-12-03 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. There is minimal atrophy involving the frontal and temporal lobes. There is no evidence of mass, lesion, hemorrhage, infarction, or metastatic disease. The substantia nigra shows normal pigmentation. The basilar cerebral vasculature shows minimal atherosclerosis. Microscopic description: The hippocampus with adjacent temporal lobe (11C), frontal lobe (2B), and pons (CBL1) are examined with the H&E stain. There is normal neuronal cellularity with no evidence of parenchymal neuropathology, hypoxia, or metastatic d1sease. The section of frontal lobe shows a small amount of subarachnoid hemosiderin. The section of pons shows a small m1ld focus of perivascular lymphocytes but is otherwise unremarkable. Comment: These findings are unremarkable and show no evidence of significant pathology. There is also no evidence of metastatic disease. The small amount of subarachnoid hemosiderin likely reflects a remote (old) resolved mild focal subarachnoid hemorrhage.
HC-3504: Neuropathological diagnosis: 1. Essentially normal brain. 2. Incidental age related changes. 3. No evidence of cerebral metastatic disease. Date Completed: 2-20-03 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined m a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. There is no apparent softening, discoloration, hemorrhage, mass, or other lesion. The junction between the cortex and white matter is well demarcated. There is no grossly identifiable evidence of metastatic disease. The striatum, lentiform nucleus, hippocampus, and thalamus are normal. There is no atrophy of the cerebellar folia. Significant atherosclerosis of the basilar cerebral vasculature is not present. Microscopic Description: A section of hippocampus (10) and adjacent temporal cortex (10) are examined with the H&E stain. There is normal neocortical and hippocampal neuronal cellularity. There is no extracellular spongiosis or gliosis. The hippocampus shows a few examples of granulovacuolar degeneration and scattered Hirano bodies. Significant formation of neurofibrillary tangles is not present. There is also no evidence of hemorrhage, necrosis, infection, or metastatic disease. Comment: These findings show incidental age-related changes.
HC-3543: Neuropathological diagnosis: 1. Essentially normal brain. Date Completed: 6-12-03 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at levels are examined in a frontal to occipital direction. Four x 1cm sections of the brainstem and cerebellum are also examined. The frontal, parietal, and temporal lobes show no atrophy. There is no evidence of mass, lesion, hemorrhage, infarction or metastatic disease. The substantia nigra shows normal pigmentation. The basilar cerebral vasculature shows no atherosclerosis. Microscopic description: The hippocampus with adjacent temporal lobe (10) is examined with the H&E stain. There is normal neuronal cellularity with no evidence of neuropathology, hypoxia, or metastatic disease. Comment: These findings are unremarkable and show no evidence of pathology.

No evidence of metastatic carcinoma.
Date Completed: 6-12-03 Gross Description: High quality, high resolution colored digital images of the entire brain and 7mm full coronal sections at all levels are examined in a frontal to occipital direction. Four I cm sections of the brainstem and cerebellum are also examined. The frontal, parietal, and temporal lobes show no atrophy. There is mild cerebral edema but no associated herniation. There is no evidence of mass, lesion, hemorrhage, infarction, or metastatic disease. The substantia nigra shows normal pigmentation. The basilar cerebral vasculature shows no atherosclerosis. Microscopic description: The hippocampus with adjacent temporal lobe (12) is examined with the H&E stain. There is normal neuronal cellularity with no evidence of neuropathology, hypoxia, or metastatic disease. To determine if developmental expression of the Plp1 i.msd allele in (A) could confer an early death phenotype equivalent to the naturally-occurring Plp1 msd severely hypomyelinating mutant mouse characterized previously 10,22,23 , we bred Plp1 i.msd mice with constitutively-expressing CNPcre mice 24 (i.e. Plp1 i.msd :: CNPcre). The Cnp gene is expressed in oligodendrocyte progenitors from mid-gestation. Wild type (WT) and single mutant allelic CNPcre or Plp1 i.msd mice do not exhibit overt phenotypes and live beyond 12mo of age (Kaplan-Meier curves are offset on the ordinate axis for visibility). Overall, the Kaplan-Meier curves are statistically different (X 2 (4) = 73.5; p < .0001), but Plp1 msd and CNPcre::Plp1 i.msd curves are indistinguishable (X 2 (1) = 2.29; p = .13), with median survival times of 24 (n = 15) and 23.4 (n = 12) days, respectively. CNPcre::Plp1 i.msd mice phenocopy the Plp1 msd mutants with respect to ataxia, spasticity, likely severe myoclonus 25 immediately prior to death and pervasive translucent pallor of white matter tracts in brain and spinal cord. F) We determined the efficiency of tamoxifen uptake by oligodendrocytes in vivo after breeding CAG-tdTomato mice (cat 007905; Jackson Laboratories, Bar Harbor, ME) with Plp1 i.msd ::McreG mice (i.e. CAG-tdTomato::Plp1 i.msd ::McreG). Cre recombination enables EGFP expression in myelinating oligodendrocytes, which we used for quantification as a proportion of total CC1 + oligodendrocytes. Tamoxifen uptake in pons and brainstem 7days after 2 x daily oral gavages shows that approximately 20% of the perivascular oligodendrocytes absorb sufficient drug for recombination of the tdTomato allele (n=1 mouse, 4 vibratome sections). This proportion decreases with distance from vessels and overall approximates 10% within 3 cell diameters and 5% overall. G) Agarose gel showing PCR amplification of genomic DNA purified from the optic nerves of 3 OBi and 3 control mice at 18mo. All mice have the strong PCR product from the wild type allele (wt), but only OBi mice have the larger product from the Cre-recombined allele (rec) which encodes PLP1 i.msd . Image J analysis of the relative intensities of the wt and rec alleles, after local background subtraction, indicates a recombination frequency of 25.4±2.5%. This is likely a steady state level reflecting recombination in cells from successive rounds of tamoxifen gavage. We cannot detect the recombinant allele in tissue from a single tamoxifen gavage using this approach. Abbreviations: PLP1/DM20, proteolipid protein 1; MBP, myelin basic protein; CNP, 2',3'-cyclic-nucleotide 3'-phosphodiesterase; a-Tub, a-Tubulin (loading control). See unprocessed western blots below. Auditory brainstem responses for control, OBi and Plp1 i.msd mice at 2 and 6mo were acquired as previously described 26 , using two channel recordings from reference (inverting) mastoid electrodes and a noninverting vertex electrode under avertin anesthesia. Interpeak latencies between waves I and V (V -I, overall PNS + CNS components), waves II and V (V -II, CNS component) and waves I and II (II -I, PNS component) as functions of sound pressure level (SPL) for 32kHz pure tone pips. Interpeak latencies are inversely proportional to SPL, and used as a noninvasive proxy of conduction velocity in myelinated auditory brainstem fibers 27,28 . The staining is generally uniform in littermate OBi mice; however, at least 2 focal demyelinated/remyelinating shadow-like plaques are visible in left stria medullaris (lower panels, arrowheads). In the event that these lesions are remyelinating, the repair is likely a function of physiologically-normal myelin synthesized by wild type oligodendrocytes because the myelinating capacity of oligodendrocytes expressing mutant proteins from the Plp1 i.msd allele is only a few percent of normal 29 . The contralateral tract in the OBi mouse and other WM tracts (arrows) are evenly stained suggesting minimal pathology. B) Internal capsule from coronal sections is weakly stained with LFB in OBi mice, reflecting bilateral generalized hypomyelination. However, we do not observe perivascular cuffing in the mutants, suggesting that infiltrating immune cells do not play a major role in the pathology out to at least 12 months.
Supplementary Information Corticohippocampal Dysfunction In The OBiden Model Of Primary Oligodendrogliopathy Daniel Z. Radecki, Elizabeth L. Johnson, Ashley K. Brown, Nicholas T. Meshkin, Shane A. Perrine and Alexander Gow Supplementary Fig. S4 -Macrophages in the brainstem corticospinal tract ingesting myelin debris in OBi mice A) Confocal image of well myelinated (EGFP from dtTomato) region (longitudinal section) from a 6mo control mouse showing a resting microglial cell (black arrowhead) labeled with antibodies against Iba-1 (red). Nuclei are labeled with DAPI (blue). B) Corticospinal tract of a 6mo OBi mouse showing a large activated microglial cell contacting a myelinated axon. The microglial cell body (left) has extended a leading process around the sheath (black arrowheads), the right hand end of which is associated with apparent myelin debris (bright green blob). In addition, the left hand end of the cell body appears to be associated with myelin debris. Further, the cell body may contain ingested myelin fragments in the cytoplasm (white arrow). C) A large activated Iba-1+ microglial cell (black arrowheads) in close proximity to or contacting another cell. D) Same field as in (C) showing that the microglial cell cytoplasm contains many CD68+ lysosomes (white arrows) and that this cell may be in the process of engulfing or attacking the adjacent cell (black arrow). Scale bar, 10µm.