Development of MCT1 BAC Reporter and MCT1 overexpressor mice
BAC MCT1 tdTomato reporter mice were produced as described previously.22
The BAC construct was modified to include the entire 19.8kb MCT1 gene plus 50kb upstream of the first exon and 132.2kb downstream of the last exon (Supplementary Fig. 2a
). Founder lines were identified by PCR (primer pair was 5′-CGAGGAGGTCATCAAAGAGT-3′ and 5′-AGAACTTGAGGTGGTCCATC-3′) and also by tdTomato fluorescence within red blood cells. Two expression lines were produced and backcrossed with B6 or C57BL/6 wild-type mice. To produce MCT1 overexpressor mice, the promoter for the astrocyte-specific gene, glial fibrillary acidic protein (GFAP), was used to drive MCT-1 overexpression specifically in astrocytes,41
was inserted between the GFAP promoter sequence and an internal ribosomal entry site (IRES)-eGFP sequence, resulting in expression of MCT-1 and eGFP together in astrocytes. Six founder lines were generated, of which five expressed MCT-1 and eGFP within CNS astrocytes.
Generation of lines coexpressing MCT1 BAC reporter and other transgenes
BAC GLT1 eGFP,22
BAC MOBP eGFP (GENSAT), PLP-eGFP (Jackson laboratory), PLP-CreER (Jackson laboratory), CNP-EGFP/Rpl10a (Jackson Laboratory), Aldh1L1 (GENSAT), Thy-I YFP (Jackson laboratory), and SOD1 G93A transgenic mice (Jackson laboratory) were crossed with BAC MCT1 reporter mice to establish double transgenic mice.
Fluorescent activated cell sorting (FACS), real-time RT PCR, RT-PCR, and BacTRAP
Brains from 1 month old BAC MCT1 tdTomato reporter mice were prepared as described previously.42
Cells were sorted into three groups with a MoFlo MLS high speed cell sorter (Beckman coulter) running Summit version 4.3 software in the FACS core at The Johns Hopkins University. Total RNA prepared from sorted cells and tissue using Absolutely RNA Miniprep Kit (Stratagene, La Jolla, CA) was converted to cDNA using cDNA synthesis kit (Applied Biosystems, Foster city, CA). PCR for MCT1 or GAPDH (control) was completed using the following primer pairs, respectively: (5′-AAAATGCCACCTGCGATTG GA-3′ and 5′-GCCTGATTAAGTGGAGCCAGG-3′), or (5′-TTGCCATCAATGACCCCTTCA -3′ and 5′-CGCCCCACTTGATTTTGGA -3′). TaqMan premade gene-specific probes and 18srRNA (as a control) were used for real time RT-PCR. For the FACS isolation of oligodendroglia and astrocytes in the brains, young adult BAC MOBP eGFP- and BAC GLT1 eGFP- mice (1–2 month) were used as described previously.42
For affinity purification of polysomal mRNAs from oligodendroglia and astrocytes in the brain, adult CNP BacTRAP (translating ribosome affinity purification)- and ALDH BacTRAP- mice were used, respectively, as described previously.43
Primary antibodies for Western blots and immunohistochemistry
Detailed protocols for Western blots and immunohistochemistry are described below. The following primary antibodies were used for Western blots: MCT1-human (AbGent; 1:75), MCT1-mouse (Santa Cruz; 1:50), MCT4 (Millipore; 1:400), CD147 (Novus Biologicals; 1:500), CNPase (Millipore; 1:1000), Connexin 43 (Millipore; 1: 500), and β-actin (Millipore; 1:1000). The following primary antibodies were used for immunofluorescence: mouse MCT-1 (Santa Cruz; 1:50), human MCT-1 (Santa Cruz; 1:50), Olig2 (Millipore; 1:500), CC-1 (Calbiochem; 1:50), CNPase (Millipore; 1:250), MBP (Covance; 1:250), PDGFRα (BD Pharmingen; 1:100), NG2 (Millipore; 1:100), GFAP (Dako; 1:2000), Iba1 (Wako; 1:250), NeuN (Millipore; 1:250), SMI-32 (Covance; 1:1000) Ubiquitin (Millipore; 1:250), Caspr (from Bergles’ lab; 1:1500), Nav1.6 (from Bergles’ lab; 1:100), DsRed (Clontech;1:250), Tuj1 (Millipore, 1:1000), NeuN (Millipore (1:1000), ALDH1L1 (Neuromab/UC Davis, 1:10), and Iba1 (Wako, 1:500).
Lactate uptake assay in oligodendroglioma and astrocytoma cell lines
MO3.13 oligodendroglioma and U87 astrocytoma cell lines are cultured as described in previous reports.44,45
Lactate uptake assay was completedas described previously46
with only small modifications. In brief, cells were incubated with 0.5 μCi/mL L-[1-14
C] lactic acid (Perkin-Elmer, Boston MA, USA) in HEPES-buffered, Earl’s balanced salt solution (HEBSS) buffer, pH 6.0, containing150mM NaCl, 5mM KCl, 1mM KH2
, 0.2mM CaCl2
O, 3.3mM MOPS, 10mM HEPES, 1mM MgSO4
O. After incubation, uptake was stopped by quickly chilling the cultures to 4 °C. Cells were washedwith ice-cold HEPES buffer, homogenized in 0.1N aOH and 0.1% Triton X100, and centrifuged at 13000rpm for 10 minutes. Radioactivity was measured byscintillation counting and corrected by protein amount.
Immunohistochemistry, histology, light and electron microscopy
Mice were anesthetized with isoflurane/oxygen and perfused transcardially with 1X PBS followed by 4% paraformaldehyde in PBS (for light microscopy [LM]) or 4% paraformaldehyde/2.5 % glutaraldehyde in PBS (for electron microscopy [EM]), tissue collected, and either cryoprotected in graded concentrations of sucrose and sectioned at 20 μm on a Leica CM1900 cryostat (for LM) or embedded in Epon resin after postfixing in osmium tetroxide (for EM).
To assess neuropathology by light microscopy, tissue sections were stained with hematoxylin and eosin (H&E), Eriochrome staining, Bielschowsky silver staining, or immunohistochemistry (IHC). For lectin staining of endothelial cells, biotinylated Lycopersicon esculentum lectin (1:200 dilution; B-1175; Vector Laboratories, Burlingame, CA, USA) was incubated with sections for 1 hour. As a secondary antibody, sections were incubated in Streptavidin-488 (1:100 dilution; B-1175; Vector Laboratories, Burlingame, CA, USA). For IHC of MCT1, tissue slices were pre-treated with sodium citrate buffer (10mM Sodium Citrate, 0.05% Tween 20, pH 6.0) for antigen retrieval (other primary antibodies did not require antigen retrieval) prior to 1 hour incubation at room temperature in blocking buffer (0.3% BSA, 5% Skim milk, and 0.3% Triton-X 100 in PBS). Primary antibodies were incubated overnight at 4°C, washed and then incubated for 2 hours at room temperature in anti-mouse or anti-rabbit IgG Alexa-fluor secondary antibodies (Invitrogen; 1:200). Photomicrographs were taken on Zeiss LSM510 Meta confocal microscopy or Zeiss Image Z1 fluorescent microscopy. Co-labeled cells with either immunostaining or transgenic reporters were manually counted in Axiovision from digital images taken by fluorescent microscopy. For EM, 1 mm thick sections were stained for toluidine blue and examined under light microscopy with 100X oil-immersion objective. 70 nm thin sections were obtained and stained for citrate/uranyl acetate. For quantification of axon degeneration in the optic nerves of MCT1 heterozygous mice and mice treated with lenti-MBP-shRNA, 3 and 10 7000x electron micrographs, respectively, were acquired for each optic nerve using a Zeiss Libra transmission electron microscope. Total number of axons and degenerating axons, as defined by degenerating myelin or dark axons, on each electron micrograph were counted and percentage axonal degeneration calculated. All quantification was completed by an investigator blinded to the genotype or treatment group.
Production of cell cultures
Oligodendrocyte cultures were produced from PND 2–3 mouse cortical tissues of PLP eGFP × NG2-tdTomato double transgenic mice. In brief, oligodendrocyte progenitor cells (OPC) were isolated by FACS and maintained in OPC culture medium (SATO medium containing PDGFα) for 4 days. OPC were further differentiated into mature oligodendroglia in the differentiation media (SATO media containing T3 without PDGFα).
Primary astrocyte cultures were produced from PND 2–3 mouse pups. Cortices were dissected out and dissociated with papain and subsequently cultured on collagen-coated T75 flask in DMEM containing 10 % fetal bovine serum (FBS). At DIV14 astroglial cells were seeded into collagen-coated 6-well plates at a concentration of 7×105 cells/well.
Organotypic or cortical spinal cord cultures
were produced from PND 7 Sprague-Dawley rat pups or MCT1 tdTomato reporter mice, as described previously.47
Anti-sense oligonucleotides or MCT1i treatment in organotypic spinal cord cultures
One week after plating organotypic spinal cord cultures, 5 μM MCT1 ASO (IDT) or various doses of MCT1i was added to media for 3 weeks. Sections were either visualized with propidium iodide (PI, 7.5μM, 2 hrs, Sigma) or fixed with 4% paraformaldehyde, and immunostained with the neuronal marker SMI-32 (Covance, 1:1000). Large, ventral horn SMI-32-positive neurons were counted by an investigator blinded to the treatment condition.
MCT1i and glucose deprivation in organotypic spinal cord cultures
Two weeks after collection, cultures were pre-treated with MCT1i27
or DMSO for 24 hrs, incubated in propidium iodide for 2 hours, and then photographed on a Nikon epifluorescence scope (Pre-treatment). Sections were then washed, incubated with glucose-free or normal buffer ± MCT1i for 2 hrs, allowed to recover for 2 hrs, re-incubated with PI for 2 hours, and photographed images quantified by fluorescent intensity (Post-treatment). To examine motoneuron loss, sections were returned to normal media for two weeks, fixed in 4% paraformaldehyde, and immunostained with SMI-32 (Covance, 1:1000). To study cellular localization of PI, sections were immediately fixed in 4% paraformaldehyde and then processed for immunocytochemistry with primary antibodies. Secondary antibodies used for detection were AlexaFluor 488-conjugated goat anti-rabbit, anti-mouse, or anti-chicken antibodies (Invitrogen, 1:1000).
Organotypic spinal cord cultures treated with bicuculline or glutamate
For bicuculline (BIC) experiment, spinal cord slice cultures (2 weeks post-culture) were treated with 100uM BIC (or vehicle) in the presence or absence of 1mM MCT1i for 3 days. To measure cell death, cultures were incubated with PI and then photographed on a Nikon epifluorescence scope. For glutamate experiment, spinal cord cultures were incubated with MCT1i (or DMSO vehicle for controls) for 24 hours, followed by media with or without 250uM glutamate buffer for 60 minute. PI staining was performed before and after the glutamate treatment to visualize dead cells. For quantification, fluorescent intensities of pre-treatment sections measured by Image J software for the dorsal horn in BIC treated sections, and total spinal cord in glutamate treated sections, were subtracted from post-treatment sections. All values were normalized to untreated control sections.
Production of lentiviral constructs
Based on pre-designed ON-TARGETplus SMARTPool siRNA sequences (Dharmacon), DNA oligos that contain MCT1-shRNA sequences were synthesized and directly subcloned into pSuper vector (Oligoengine, Inc.). The sense and antisense oligo sequences for the most efficient MCT1-shRNA were as follows: 5′-GATCCCC GTATCATGCTTTACGATTATTCAAGAGATAATCGTAAAGCATGATACTTTTTTC-3′ 5′-TCGAGAAAAAAGTATCATGCTTTACGATTATCTCTTGAATAATCGTAAAGCATGATACGGG-3′ DNA fragments spanning the H-1 promoter and the shRNA sequences in pSuper construct were amplified by PCR and subcloned into FUGW-CMV to produce Lenti-shRNA. Fragments spanning MBP promoter, shRNA, and IRES-EGFP sequences were subcloned into FUGW-CMV to produce Lenti-MBP-shRNA. MCT1 shRNA was subcloned into the HpaI and XhoI restriction sites of pSico (Addgene) to produce Cre-inducible lentivirus. Lentiviral constucts, including control GFP constructs, were produced in HEK 293 T cells using the FUGW-CMV/Δ8.9/VSVG system.
MCT1-shRNA Lentivirus injected into spinal cord
Unilateral injections of LV-MCT1shRNA (n = 10) or LV-GFP (n = 8) were performed in C57BL6 wild-type mice at 100 days-of-age at C4, C5, and C6, delivering 1.7 × 105 PFU for LV-GFP or 1.8 × 105 PFU for LV-MCT1shRNA per mouse. Briefly, the transverse processes of C4 to C6 were removed, the dura removed, and three 1 μl injections of lentivirus or media were injected with a 34 gauge Hamilton syringe needle. Animals were sacrificed after 4 weeks by transcardial perfusion with 4% paraformaldehyde. Spinal cord sections were cryoprotected, sectioned on a cryostat, and immunostained for neurofilaments (SMI-32), oligodendrocyte-lineage cells (Olig2), microglia (Iba1), or astrocytes (GFAP), as described above. Large (diameter greater than 20 um) neurofilament-containing neurons in the ventral spinal cord were counted from every third section that localized GFP to the ventral horns on both the virus- and media-injected halves of the spinal cord (mean number of sections counted 11.9 for shRNA group and 12 for GFP group). For each animal, the mean number of motoneurons per section from the virus-injected spinal cord hemisection was divided by the mean number of motoneurons per section from the contralateral media-injected motor neurons per section.
Generation of MCT1 heterozygote null mice
MCT1 heterozygote null mice were generated by targeted homologous recombination to replace a 640 base pair sequence of the MCT1 gene starting at the translation initiation codon and containing exon 1 as well as part of the first intron with the LacZ gene sequence fused with a neomycin (Neo) resistance gene sequence and put in frame with the MCT1 promoter (S. Lengacher et al., manuscript in preparation). Successful recombination event and proper insertion of the LacZ/Neo sequence in the targeted locus was controlled by Southern blot on DNA from embryonic stem cells previously electroporated with the targeted vector and selected with gancyclovir and G418. Genotyping of animals was performed by PCR with appropriate set of primers.
MCT1 shRNA driven by MBP promoter injected into rat optic nerves
Sixty day old Sprague Dawley rats were anesthetized with 2 % isofluorane, an incision placed in the scalp overlying the orbital ridge, the skin overlying the orbital ridge retracted, and a conjunctival suture placed to provide traction to the globe. The optic nerve was localized, a window cut in the overlying dura, the nerve pierced with a pulled micropipette, and each nerve infused with 3 μl of either lenti-MBP-shRNA or lenti-MBP-GFP (2.4 × 1011 virus particles (VP)/ml). After 4 weeks, the rats were transcardially perfused with 4% paraformaldehyde and the optic nerves from the globe to the optic chiasm carefully dissected. The optic nerve near the injection site was post-fixed in 4% paraformaldeyde, cryoprotected in 25% sucrose and cut on a cryostat. These nerve segments were later immunostained with Olig2, GFAP, and SMI32. Optic nerve distal to the injection was post-fixed in 4% paraformaldehyde/2.5% glutaraldehyde for 3 days and then processed for EM as described above.
Cre-dependent MCT1 shRNA Lentivirus injected into corpus callosum
Cre-inducible MCT1 shRNA lentivirus was injected bilaterally into the corpus callosum of PLP-CreER,36
MCT1 tdTomato reporter, and wild-type mice at approximately 60 days of age. Mice were anesthetized with ketamine and xylazine, placed in a stereotaxic device, and the corpus callosum injected bilaterally with 2 μl lentivirus (2.6 × 1011
VP/ml), as previously described.48
After 1 week, mice were injected intraperitoneally with 1 mg tamoxifen for 5 consecutive days to induce Cre recombination of nucleus-integrated virus. Mice were sacrificed 4 weeks later by transcardial perfusion with 4% paraformaldehyde. Sections were cryoprotected, sectioned on a cryostat, and immunostained for neurofilaments (SMI-32) and green fluorescent protein (GFP), as described above.
Western blotting of human autopsy samples, primary cells and mouse tissues
Autopsy samples were obtained from the Johns Hopkins University Brain Resource Center and the Johns Hopkins ALS Tissue Bank. Human or mouse samples were homogenized in TBS with 0.1% SDS, 1% triton and 10% glycerol, primary cultures were lysed as described previously.49
Western blotting and densitometry was performed using the enhanced chemilluminescence (ECL) system (GE Healthcare) and NIH Image J.
In-vitro ASO and MCT1i experiments were analyzed using one-way ANOVA followed by Tukey-Kramer post hoc test. Western blots, real-time RT PCR, and in-vivo LV-MCT1shRNA quantifications were analyzed using unpaired Student’s t-tests. Quantification of axon degeneration in lentivirus injections of optic nerve was evaluated with a Mann Whitney test.