Isolation, Purification and Verification of Peripheral Nerve Myelin Derived from Bovine Cauda Equina

Copyright: © 2012 Xia RH, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction
Myelin, the complex biological membrane that facilitates saltatory transmission of electrical impulses along axons, is derived from the differentiation of Schwann cell plasma membranes in the peripheral nervous system (PNS) and from oligodendrocytes in the central nervous system (CNS) [1]. Although similarities exist between PNS and CNS myelin, there are unique differences in molecular composition. Extensive information is known on myelin composition, synthesis and maintenance in mammalian and non-mammalian species [1]. The importance of myelin to PNS function is evidenced by the devastating effects of demyelination in acquired autoimmune neuropathies such as Guillain-Barré syndrome (GBS) and Chronic Inflammatory Demyelinating Polyradiculoneuropathy and toxic neuropathies, such as due to perhexiline [2]. Studies using native myelin remain important despite advances in peptide biosynthesis. This is the case for a severe murine experimental autoimmune neuritis (EAN) model of GBS [3][4][5][6]. Mice immunized with purified myelin proteins or peptide fractions are often resistant or develop very mild forms of the disease, making these models inadequate to study PNS inflammation and demyelination seen in GBS [7]. Similarities in neural-immune responses between mice and humans make it essential to work with reliable mouse PNS inflammation models to understand the pathogenesis of demyelinating polyneuropathies and find novel molecular therapeutic targets. Detailed protocols using modern equipment and consumables to isolate mammalian PNS myelin are difficult to find in the medical literature. We adapted or modified traditional protocols to isolate, purify and rapidly verify bovine PNS myelin and its protein and lipid components using readily available resources available to most general scientific laboratories. Myelin isolates were used to reliably induce EAN in susceptible female mice.

Isolation and purification of bovine peripheral nerve myelin from cauda equina
Bovine peripheral nerve myelin (BPNM) was isolated from frozen intrathecal cauda equina of United States Department of Agriculturecertified adult cows <30 months of age (Innovative Research, Novi, Michigan, U.S.A.), with modifications [8,9]. The thawed cauda equina was carefully dissected and ground to a fine powder in liquid nitrogen and weighed. A 5% weight/volume solution in 0.29 M sucrose (>99.9% pure; EMD Chemicals, Gibbstown, New Jersey, U.S.A.) in distilled water was homogenized three times at 25,000 revolutions per minute for at least 30 seconds on ice using a Polytron PT 1200E hand-held homogenizer (Brinkmann-Kinematica, Lucern, Switzerland). The homogenate was filtered through a 50 µm Nitex ® nylon lab pak mesh (Sefar Filtration Inc., Depew, New York, U.S.A.) and kept overnight at 4°C.
A discontinuous sucrose gradient was generated by layering the filtered homogenate over a 0.85 M sucrose solution in distilled water into an Optiseal TM polyallomer ultracentrifuge tube (Beckman Coulter, Palo Alto, California, U.S.A.) in a 3:4 ratio. The gradient was centrifuged at 82,500 g for 45 minutes at 4°C in an Optima LE-80K Ultracentrifuge with Type 70 Ti fixed angle rotor (Beckman Coulter). The crude myelin was carefully collected from the sucrose interface using Luer Lok TM tip syringes attached to 15G BD Yale TM reusable hypodermic needles (BD Labware, Franklin Lakes, New Jersey, U.S.A.), and subjected to osmotic shock four times by resuspending 1:3 in distilled water, homogenizing and centrifuging at 82,500 g for 15 minutes at 4°C as described above. The supernatant (containing cellular fragments and debris) was completely removed and the myelin pellet completely aspirated. The partly purified myelin extracts were pooled and stored at 4°C overnight.
Further purification was performed using a continuous cesium chloride (99.999% pure; MP Biomedicals, Solon, Ohio, U.S.A.) gradient, generating a 0.3 M cesium chloride-myelin suspension in distilled water. A continuous 0.3-1.3 M cesium chloride gradient was generated (equal volume of 1.3 M cesium chloride in distilled water added to the bottom of a 0.3 M solution in Optiseal TM ultracentrifuge tubes, carefully titled to 45°, rotated 360° across their vertical axes then brought upright within 60 seconds). The 0.3 M cesium chloride-myelin suspension was layered over the 0.3-1.3 M continuous cesium chloride gradient (in a 1:5 ratio) and centrifuged at 82,500g for 30 minutes at 4°C. Purified myelin was collected and subjected to osmotic shock as described above. Purified myelin was lyophilized using a RC 10

Transmission electron microscopy (TEM)
Negatively-stained myelin images were acquired as previously described, with modifications [10]. Briefly, a 5 µL droplet of BPNM (100-200 µg/mL in deionized distilled water) was applied to a Formvarcoated, carbon stabilized 200 mesh copper grid for one minute. The droplet was then wicked away with Whatman ® filter paper (No.1; Maidstone, England) and replaced immediately with a droplet of 1% aqueous potassium phosphotungstate for one minute, after which Whatman filter paper was again used to wick the solution away. The grid was allowed to air dry for a few minutes before viewing on the electron microscope. Photomicrographs were recorded on a JEOL 200CX electron microscope (Tokyo, Japan) operating at 80 KeV. All photomicrographs were digitized by scanning the negatives into an

Light and phase contrast microscopy
Light microscopy of BPNM smears was performed by adapting a protocol for visualizing intact myelin sheaths in peripheral nerves [11]. 40-50 µL of a 20 mg/mL suspension of lyophilized BPNM in deionized distilled water was layered evenly unto poly-l-lysine coated glass slides (VWR Scientific) and incubated for 10 minutes at 37°C to bind myelin. Smears were fixed with 100 µL of 3% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4 at RT for 30 minutes in a fume hood. BPNM smears were post-fixed with 1% osmium tetroxide in 0.1 M phosphate buffer and stained with 0.1% toluidine blue in distilled water, washed, mounted, sealed with quick drying clear nail polish and viewed immediately using a Zeiss Axioskop Epifluorescent Microscope (San Diego, California, U.S.A.). Digital images were generated using a Zeiss Axiocam MRc 5 digital camera and processed with the Adobe Photoshop CS2 software program (Adobe Systems, San Jose, California, U.S.A.). For BPNM phase contrast microscopy, smears were prepared as described above and fixed with 100 µL of 4% paraformaldehyde (PFA) in 1X PBS for 5-10 minutes at 37°C. After washing three times with 100 µL of fresh 1X PBS for 5 minutes each, glass coverslips were mounted on the slide with a small drop of Vectashield ® (Vector Laboratories, Burlingame, California, U.S.A.), sealed, viewed and imaged as described above.

Gravimetric analysis
BPNM fractionation into lipid and protein components for gravimetric analysis was performed by adapting previously published protocols [12,13]. 5-10 mg of lyophilized BPNM was weighed with a Sartorius CPA124S analytical balance (Goettingen, Germany) and mixed thoroughly with 4 volumes of distilled water. 20 volumes of 2:1 volume/volume chloroform (CHCl 3 ): methanol (CH 3 OH) were added, mixed thoroughly at RT and filtered using pre-weighed Whatman ® ashless analytical filter paper (No. 42; Maidstone, England). The insoluble residue was filtered three times with 20 volumes of 2:1 CHCl 3 :CH 3 OH and pooled for subsequent use. The residue was separately washed with the same volume of water, 100% ethanol and 100% ether on filter paper that was carefully dried in nitrogen gas (Alphagaz 1 N2 99.999% pure, Air Liquide, Houston, TX, U.S.A.) and weighed using an analytical balance to determine the amount of insoluble residue.
To separate lipids and proteolipid-protein fractions, the volume of the pooled filtrate in a single borosilicate glass tube was measured. A 0.2 volume of 0.05 M potassium chloride in deionized distilled water was added and mixed thoroughly to generate a single phase, and centrifuged with a Beckman Coulter Allegra X-12R Centrifuge (Fullerton, California, U.S.A.) in a swinging bucket rotor at 2400 revolutions per minute for 20 minutes at RT. The lower phase was collected and saved. The tubes were washed twice with small amounts of upper phase solvent (CHCl 3 :CH 3 OH:H 2 O in ratio 3:48:47) and centrifuged as above to completely remove the lower phase. The total volume of the lower phase was measured, 100% CH 3 OH was added (contains CHCl 3 : CH 3 OH: H 2 O in ratio 86:14:1) to restore 2:1 CHCl 3 :CH 3 OH ratio and dried under nitrogen stream at 60°C.
Four volumes of distilled water and 20 volumes of 2:1 CHCl 3 :CH 3 OH were added to the dried substance, resuspended, and centrifuged at 2400 revolutions per minute for 20 minutes at RT as above. The lipid containing supernatant was collected and transferred to a clean, previously weighed glass tube, leaving the precipitate (containing proteolipid-protein and neutral solvent insoluble lipid) behind. The process was repeated at least 6 times until there was no visible precipitate. The precipitate was pooled in a single previously weighed glass tube and resuspended in at least 20 volumes of 2:1 CHCl 3 :CH 3 OH. The suspension was centrifuged as above, and heated at 65°C for 20-30 minutes to extract neutral solvent insoluble lipid from the pellet (maximal constriction of the pellet being indicative of complete extraction). The cooled supernatant was transferred to the lipid containing supernatant, dried completely to constant weight under nitrogen stream, then in vacuo with Drierite and paraffin oil placed in a Nalgene Nunc vacuum dessicator (Thermo Fisher Scientific, Rochester, New York, U.S.A.). Similarly, the precipitate was dried under nitrogen stream and in vacuo under Drierite to constant weight, yielding the amount of the proteolipid-protein fraction.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
BPNM was partially delipidated by treatment with ether-ethanol (3:2, volume/volume), adapting a previously published protocol [8]. Briefly, 5 mg of lyophilized BPNM suspended in the solvent (final concentration of 1 mg/mL) was mixed vigorously. The protein pellet was collected by centrifugation at 2500 g for 3 minutes at RT in a Beckman Coulter Allegra X-12R Centrifuge, repeating 5 times. The final pellet was dried under nitrogen stream at RT, resuspended in deionized distilled water and homogenized using a Branson sonifer 250 sonicator (VWR Scientific). Protein concentration was determined by absorbance at 280 nm with a BMG Labtech FluoStar Optima microplate reader (Durham, North Carolina, U.S.A.), using known concentrations of bovine serum albumin (BSA; Innovative research) to generate standard curves. Protein samples were stored at -80°C prior to electrophoresis.
The Mini-Protean III apparatus (Bio-Rad Laboratories, Richmond, California, U.S.A.) was used for gel electrophoresis, according to the manufacturer's instructions, with 15% polyacrylamide slab and stacking gels. Denatured 50 µg aliquots of BPNM protein, 15µg bovine brain myelin basic protein (MBP; 50% pure, Sigma-Aldrich) and BSA standards (1.0-25.0 µg) were used for electrophoresis. The Precision Plus Kaleidoscope molecular weight marker (Bio-Rad Laboratories) was used for molecular weight determination. BPNM protein subfractions were identified based on expected molecular weight. Bovine brain MBP was used as a control to verify that the isolated myelin was not of central origin. Electrophoresis was run for 50-60 minutes at 175 V.
Following electrophoresis, the gel was carefully washed and stained with 0.1% weight/volume Coomassie brilliant blue G (Sigma-Aldrich) mixed with 25% methanol (volume/ volume) and 5% glacial acetic acid in deionized distilled water for 1 hour at RT in a Sci-ERA orbital shaker (Bellco Glass Inc., Vineland, New Jersey, U.S.A.). Digital images were obtained using an AlphaImager ® HP gel documentation system (Cell Biosciences, Santa Clara, California, U.S.A.) attached to a Sony ICX267AL 1.39 Megapixel CCD camera and processed using the AlphaView and Adobe Photoshop CS2 software programs. BPNM protein subfractions were quantified by spot densitometry, using known quantities of BSA to generate standard curves. The percentage of each subfraction relative to the total amount of myelin protein quantified on the gel was determined.

Lens culinaris agglutinin fluorescent western blot
Glycoprotein detection was performed using fluoresceinated lens culinaris agglutinin (LCA-FITC; Vector Laboratories, Burlingame, California, U.S.A.) following SDS-PAGE, as described [14] with modifications. LCA is a plant lectin that specifically binds to glycoprotein α-mannose residues. Protein subfractions were transferred from the gel to polyvinylidene difluoride membranes using a Bio-Rad Mini Trans-Blot Electrophoretic Transfer Cell apparatus, according to the manufacturer's instructions. Membranes were blocked for 30 minutes at RT with 3% BSA in Tris-buffered saline (TBS). 10 µg/mL LCA-FITC in 1% BSA and 0.05% Tween-20 in TBS was applied to the membranes for 1 hour at RT in the dark in an orbital shaker. Following three washes in 0.5% Tween-20 in 1X phosphate buffered saline (PBS), the membrane was excited at 365 nm, visualized and imaged with the AlphaImager ® HP system (Cell Biosciences, Santa Clara, California, U.S.A.) using appropriate filters. BSA standards acted as negative glycoprotein controls.

Indirect fluorescent immunohistochemistry
Monogalactosylsphingolipids are a major component of myelin, a feature that differentiates this from other biological membranes [1]. BPNM was layered on poly-l-lysine coated glass slides, fixed with 4% PFA, washed three times with 100 µL of fresh 1X PBS for 5 minutes per wash, and blocked with 100 µL of 1% normal goat serum (NGS) in 1X PBS for 60 minutes at RT in a humid hybridization bath chamber (VWR Scientific). Without washing, slides were incubated with 100 µL of freshly prepared 5 µg/mL mouse anti-cow galactocerebroside IgG 3 antibody (Millipore Corporation, Billerica, Massachusetts, U.S.A.) diluted in 1% NGS in 1X PBS for 60 minutes at RT. Following three washes, slides were incubated with 100 µL of freshly prepared 1:100 dilution of goat anti-mouse IgG F(ab') 2 -FITC (SouthernBiotech, Birmingham, Alabama, U.S.A.) for 60 minutes at RT in a humid chamber in the dark. After washing with 1X PBS as above, coverslips were mounted in Vectashield ® , sealed, visualized and imaged digitally as previously described. Sulfatide O4 detection was performed as described above, using 5% fetal calf serum for blocking, 20 µg/mL mouse anti-cow O4 IgM antibody (Millipore Corporation) diluted in 1X PBS as primary antibody and 1:250 dilution of goat anti-mouse IgM-FITC (SouthernBiotech) as secondary antibody. Brief 30-60 second washes with 100 µL of fresh 1X PBS with gentle horizontal rotation were performed before and after secondary antibody incubation.

Lipid analysis
Lipids were separated from 10 mg of lyophilized BPNM modifying an established protocol [12], as described in the gravimetric analysis section. BPNM lipid extracts were evaporated to dryness under nitrogen stream at RT, and redissolved in 3-4 mL 2:1 CHCl 3 : CH 3 OH for analysis. Detection and quantification of the major myelin lipid subclasses was performed using colorimetric assays adapted from previously described protocols [15]. All the reagents used were purchased from Sigma-Aldrich, St. Louis, Missouri, U.S.A. unless otherwise stated.
For phospholipid analysis, all glassware was washed with 98% concentrated sulfuric acid, rinsed in distilled water followed by methanol and chloroform and used immediately. A 1 mL aliquot of BPNM lipid was placed in a 50 mL glass beaker and dried completely under a nitrogen stream at 35°C on a VWR hotplate/stirrer in a fume hood. 2 mL of 70% perchloric acid (Alfar Aesar, Ward Hill, Massachusetts, U.S.A.) and soda lime glass beads (VWR Scientific) were added, and heated to an internal temperature of ~110°C to release inorganic phosphorus for at least 30 minutes (until digest was clear and colorless). The digest was diluted with distilled water (total volume 12.5 mL). 2.0 mL of freshly prepared and filtered 1% weight/volume 2, 4-diaminophenol dihydrochloride with 20% weight/volume sodium metabisulfite in distilled water was added and mixed, followed by 1.0 mL 8.4% weight/volume ammonium molybdate solution in distilled water. After 20 minutes, the solution was diluted to a final volume of 25 mL. 2 mL aliquots were placed in 24-well tissue culture treated polystyrene microplates (BD Labware, Franklin Lakes, New Jersey, U.S.A.). Absorbance was measured at 620 nm in quadruplicate using a BMG Labtech FluoStar Optima microplate reader relative to reagent blank wells. A standard curve using 0-100 µg phosphorus/ mL was generated with KH 2 PO 4 , and read simultaneously with BPNM lipid phosphorus. The amount of phospholipid in the initial aliquot was calculated based on a molecular weight of 775 [13].
For galactolipid analysis, a dried 1 mL aliquot of BPNM lipid was resuspended in 2.0 mL of distilled water in a 50 mL glass beaker. 1.0 mL of 5% phenol solution in distilled water was added and mixed thoroughly, followed by 5.0 mL of concentrated sulfuric acid. The mixture was placed in a water bath at 80°C for 10 minutes. After cooling for 20 minutes at RT, 2 mL aliquots were placed in 24-well microplates and absorbance measured against reagent blanks at 490 nm in triplicate. A standard curve of 0-80 µg galactose/ mL was generated and read simultaneously with BPNM lipid. A molecular weight of 846 was used to determine the amount of galactolipid in the initial aliquot [13].
For total cholesterol analysis, a dried 1 mL aliquot of BPNM lipid was redissolved in 6.0 mL of glacial acetic acid in a 50 mL glass beaker. A stock solution of 2.5% weight/volume FeCl 3 .6H 2 O in 85% H 3 PO 4 was diluted in a 2:23 ratio with concentrated sulfuric acid, and 4 mL of this solution added to the glass beaker and mixed thoroughly. After cooling for 10 minutes at RT, 2 mL aliquots were placed in 24-well microplates and absorbance measured at 544 nm against reagent blanks in quadruplicate. A standard curve of 0-0.5 mg/mL cholesterol (>99% pure, Sigma-Aldrich) was generated and read at the same time as BPNM lipid. The total amount of myelin lipid in each extract was calculated by adding the mean amounts of phospholipid, galactolipid and cholesterol, and multiplying by the total volume of 2:1 CHCl 3 :CH 3 OH solvent used after initial lipid separation. The percentage of each lipid subfraction relative to total lipid dry weight was calculated for each extract.

Structural characteristics of isolated BPNM
Cauda equina derived BPNM was verified by negatively stained TEM, a method designed to detect subcellular fractions of animal tissues [10]. The characteristic repetitive lamella pattern of myelin was seen. The lack of debris or cellular organelles was observed at several magnifications ( Figures 1A-1C). Furthermore, light microscopy of glutaraldehyde-fixed, osmium tetroxide post-fixed BPNM smears counter-stained with 0.1% toluidine blue showed aggregates of myelin ( Figures 1D and 1E). Gravimetric analyses demonstrated that lipids accounted for ~72% of the dry mass of BPNM, while ~25% was protein/ proteolipid and <4% insoluble residue ( Figure 1F). These values are consistent with the known composition of myelin [1,13], verifying the isolation and purification process.

Protein composition of purified BPNM
SDS-PAGE demonstrated that BPNM was made up of 5 major protein fractions with molecular weights ~15 KDa, ~19 KDa, ~23 KDa, ~28 KDa and ~170 KDa. These are consistent with myelin proteins P2, MBP, peripheral myelin protein-22 (PMP-22), myelin protein zero (P0) and periaxin [1] as shown on Figure 2A. It is also possible that the 19 and 23 KDa bands represent or contain in part, in situ proteolytic fragments of P0 [16][17][18]. Fluorescent western blotting using LCA-FITC confirmed that BPNM consisted of predominantly glycoproteins [1], with partial staining of the MBP band indicating non-specific binding or contamination with a P0 proteolytic fragment. P2, a fatty acid binding-basic protein, was not detected, neither were the BSA standards, indicative of selective glycoprotein detection by LCA-FITC ( Figure 2B). Based on quantitative band spot densitometry, Values represent percentages of proteins relative to total BPNM protein quantified on gels by spot densitometry (to 2 decimal points).  P0 accounted for ~59% of total bovine myelin protein, while MBP (~23%), P2 (~13%), periaxin (~4%) and PMP-22 (~1%) made up the other fractions detected (Table 1). These data verified the PNS origin of purified BPNM isolates.  Figures 3A and 3D) clearly show aggregates of myelin that correspond to areas of immunoreactivity ( Figures 3B and 3E). A partial phase contrast-fluorescent image composite confirmed the localization of galactocerebroside ( Figure 3C) and sulfatide ( Figure 3F) immunoreactivity on the myelin membranes. Phospholipids accounted for ~60% of the total dry weight of BPNM, while galactolipids and cholesterol accounted for ~24% and ~15% respectively ( Table 2). These values are within the range expected for PNS myelin [1], with a reversal in percentages of galactolipids and cholesterol when compared to a previously described analysis of bovine cauda equina myelin [19]. These relatively straightforward colorimetric assays verified the PNS origin of the isolated myelin without need for chromatography.

Histopathology of BPNM-induced EAN in female SJL/J mice
Toluidine blue stained, basic fuchsin counterstained semi-thin plastic embedded sections demonstrated multi-focal demyelination with frankly demyelinated or thinly myelinated axons and reduced large myelinated axonal density associated with mononuclear cell infiltration at peak disease severity ( Figure 4A). Indirect immunohistochemistry confirmed loss of the normal honeycomb myelin architecture due to loss or damage to Schwann cell membranes ( Figure 4B) and reduced axonal density ( Figure 4C) associated with intense mononuclear cell infiltrates. These infiltrates were predominantly macrophages ( Figure 4D), followed by T-lymphocytes ( Figure 4E) with scattered B-lymphocytes ( Figure 4F), consistent with a representative animal model of Guillain-Barré syndrome, as previously characterized and published [3][4][5][6].

Discussion
PNS myelin is still considered a valuable resource for studying inherited and acquired causes of peripheral nerve disease. The traditional methods of isolating myelin from mammalian tissues were initially described ~40-50 years ago, and utilize equipment and consumables that have been replaced or are no longer in use. Prior verification of myelin isolates required labor-intensive gravimetric analyses, SDS-PAGE and chromatography, often requiring the production of large quantities of myelin. The purpose of this study was to update these protocols and demonstrate several simplified methods of verifying the nature and purity of myelin isolates, utilizing widely available materials. This allows small scale rapid isolation, purification and verification of myelin isolates in the laboratory, prior to performing more labor-intensive and expensive analytical assays.
Negatively-stained TEM is the quickest way to accurately verify myelin and demonstrate lack of contamination. Light and phase contrast microscopy of BPNM smears can be used in situations where TEM is not readily available. An inability to detect galactocerebroside and sulfatide by indirect fluorescent immunohistochemistry could Values represent percentages of lipid subclass relative to total BPNM lipid dry weight (to 1 decimal point). imply galactolipid loss or isolation of another biological membrane. Gravimetric analytical procedures, although more labor-intensive, can provide more accurate determination of the lipid and protein components of isolated myelin. PNS myelin consists of 70-80% lipid and 20-30% protein by dry mass [1], so this method can rapidly verify that the extract is myelin prior to more detailed analyses.
We determined the protein components of BPNM based on their known molecular weights. Specific antibodies against bovine PNS and CNS myelin proteins are not commercially available or are difficult to generate. This limits the precise identification of protein components by western blot. The electrophoretic pattern and relative percentages of the protein components demonstrate that the myelin extracts were from the PNS, with P0 being the most abundant protein [1]. This difference is borne out using a commercially available 50% pure bovine brain MBP as a control in these experiments. Bovine brain MBP consists of several components that vary from 17-18.5 KDa in size. MBP may undergo post-translational modification resulting in slightly larger band sizes . Due to the lack of specific antibodies against bovine myelin proteins, glycoprotein detection using LCA-FITC provides a rapid and inexpensive method to verify PNS myelin by fluorescent western blot. Glycoproteins are major constituents of PNS myelin and minor constituents of CNS myelin [1]. Where available, CNS myelin from the same animal source could be utilized as negative controls in these assays.
Colorimetric assays can be employed to estimate the relative proportions of the major myelin lipid subclasses. These values can be compared with data derived from previous studies or other species. In this study, the relative proportions of galactolipids and cholesterol in BPNM were reversed compared to a published report describing the isolation, purification and analysis of BPNM [19]. In that study, myelin was purified using a discontinuous sucrose gradient. We performed an additional continuous cesium chloride gradient to further purify myelin isolates, as large amounts of insoluble residues were reportedly obtained in some of their isolates [19].
We used rapid and relatively inexpensive colorimetric assays to quantify the major lipid subcomponents and indirect immunohistochemistry to detect galactolipids on intact myelin preparations; in contrast to chromatography employed in the older study [19]. Furthermore, that study did not analyze BPNM protein components. Thus, our study provides important information on BPNM isolates from the cauda equina utilizing protocols with readily available materials to the general laboratory scientist. Differences in myelin purification and lipid analysis methods could explain the observed variation in galactolipid and cholesterol. It is also possible that the age, strain, dietary changes and genetic modifications of cows from which the cauda equina were obtained are contributory factors.
BPNM extracted and verified using our protocols induced a severe inflammatory polyneuritis in female SJL/J mice that recapitulates essential features of the acute inflammatory demyelinating polyradiculoneuropathy variant of GBS. This model is reliable and reproducible with ~100% induction rates [3][4][5][6], providing a means to study molecular determinants and signaling pathways implicated in acute peripheral nerve inflammation with potential therapeutic implications in humans. Mouse models of peripheral nerve inflammation more closely resemble the human disorder than rat models due to similarities in neural-immune responses. In summary, we updated and developed adapted protocols for the isolation, purification and verification of BPNM, using tools readily available to most biomedical research laboratories in industrialized nations. These techniques can be readily applied by neuroscientists, immunologists, biochemists and molecular biologists interested in studying the structure-function relationships of myelin and other biological membranes, or animal models of peripheral neuroinflammation.