Cell lines and reagents
We maintained the BV-2 mouse microglia cell line (kind gift of Dr. Sanjay B. Maggirwar) in 10% fetal bovine serum (FBS) (Atlas, Fort Collins, CO, USA, F-0500-A) in (D)MEM (Cellgro, Manassas, VA, USA 15-013-CV) with 2 mM GlutaMax (Gibco, Grand Island, NY, USA, 35050–061) and penicillin-streptomycin (Gibco, 15140–122) at 37°C at 5% CO2. We dissected 18-day embryonic rat pups to obtain hippocampal neurons for our microfluidic co-culture experiments. We plated and maintained primary hippocampal neurons in Neurobasal media (Gibco, 21103–049) supplemented with 5% FBS, B27 supplement (Gibco, 17504–044), 2 mM GlutaMax, and 50uM L-glutamic acid (Sigma, St. Louis, MO, USA, G5889-100G). We resuspended and plated BV-2 cells in the same supplemented Neurobasal media for all co-culture experiments. We obtained full length HIV-1 Tat101 from Philip Ray (University of Kentucky) that we used for all experiments involving Tat exposure. We purchased the small molecule LRRK2 kinase inhibitor LRRK2-IN-1 (LRRK2i) from Millipore (Billerica, MA, USA, 438193, 1 μM working concentration) and purified AnnexinV from eBioscience (San Diego, CA, USA BMS306, working concentration 1 μg/ml). Unless otherwise noted, we plated BV-2 cells in poly-D-lysine (Sigma, P1149-100 mg) coated 12-well plates or on glass coverslips without penicillin-streptomycin overnight and treated as indicated with 1 μg/ml Tat101, saline vehicle control, dimethyl sulfoxide (DMSO) vehicle control, or 1 μM LRRK2i for 12 hours, as indicated by our preliminary time-course experiments (data not shown). For BV-2 cell counts, we used a hemocytometer with 0.4% trypan blue stain (Gibco, 15250).
We harvested cells by scraping into 1X PBS and centrifuging at 7,000 rpm for 1 minute. We resuspended the pellet in cell lysis buffer with protease inhibitors and lysed it on ice with periodic vortexing for 30 minutes. We normalized protein concentration by Bradford assay (BioRad, Hercules, CA, USA 500–0113, 500–0114, 500–0115). We mixed 10 μg of sample with loading dye, heated it at 70°C for 5 minutes, and ran it on a 4% to 15% SDS-PAGE gel (BioRad, 456–1086) at 100 V for 1 hour. We transferred the gel onto a polyvinylidene difluoride (PVDF) membrane at 100 V for 1 hour on ice. We blocked membranes in 5% milk in 1X Tris-buffered saline (TBS) for 1 hour at room temperature with shaking. We washed membranes 3 times in 1X Tris-buffered saline with 0.1% Triton-X 100 (TBST) . We applied primary antibodies overnight at 4°C with shaking in 5% milk in 1X TBST at the following concentrations; rabbit anti-LRRK2 (1:1000) (Epitomics, Burlingame, CA, USA, 3514–1), rabbit anti-pS935-LRRK2 (1:1000) (Epitomics, 5099–1), β-actin (1:500) (Santa Cruz, Santa Cruz, CA, USA, sc-47778) . We washed membranes 3 times in 1X TBST. We applied horseradish peroxidase (HRP)-conjugated secondary antibody (GE Healthcare, Pittsburgh, PA, USA ) at a concentration of 1:5,000 in 5% milk in 1X TBST for 1 hour at room temperature with shaking. We washed membranes and applied enhanced chemiluminescence (ECL) substrate (Pierce, Rockford, IL, USA, 32106) for 1 minute. We exposed and developed membranes on film (Kodak, Rochester, NY, USA, 111–1681).
We isolated RNA from BV-2 cells using the RNeasy RNA isolation kit (Qiagen, Valencia, CA, USA, 74104). We synthesized cDNA from 0.5 μg of total RNA using the SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, CA, USA, 18080–051). We used the TaqMan Universal PCR Master Mix (Invitrogen, 4304437) with Invitrogen TaqMan Gene Expression Assay primers and probed for TNF-α (Mm00443258_m1), IL-6 (Mm00446190_m1), MCP-1 (Mm00441242_m1), IL-10 (Mm00439614_m1), IL-4 (Mm00445259_m1), BAI1 (Mm01195143_m1), Tim4 (Mm00724709_m1) and 18S ribosomal RNA (Mm03928990_g1) as an internal control. We ran and measured the samples on an Applied Biosystems (Carlsbad, CA, USA) StepOnePlus real time PCR machine. We analyzed relative target gene levels by comparing the fold change of delta-delta threshold cycle to the control after normalization to 18S ribosomal RNA.
We adapted our immunocytochemistry (ICC) protocol from Glynn et al.. Briefly, we fixed the cells in a solution of 4% paraformaldehyde, 4% sucrose in 1x PBS at 4C for 10 minutes. We then treated the fixed cells with 100 mM glycine in 1x PBS and washed in 1x PBS for 5 minutes. We prepared the primary antibodies in 3% BSA (Sigma, A3294-50 G) in 1x PBS at the following concentrations: rat anti CD11b 1:200 (Serotech, MCA711), mouse anti Tau5 1:500 (Calbiochem, 577801), rabbit anti Synapsin-1 1:500 (Cell Signaling, 5297S), and rabbit anti LRRK2 1:500 (Novus, St. Charles, MO, USA, NB300-268). We incubated in the fixed cells in the primary antibody solution for 1 hour, then washed 4 times with 3% BSA, 1x PBS for five minutes apiece. We prepared alexa fluor secondary antibodies (Molecular Probes, Eugene, OR, USA) in 3% BSA, 1x PBS at 1:500, and incubated the cells in secondary antibody for 30 minutes. We then washed the cells 4 times in 1x PBS for five minutes apiece, before mounting them on glass slides using the Prolong Gold with 4',6-diamidino-2-phenylindole (DAPI) mounting agent (Life Technologies, Carlsbad, CA, USA, P36935).
Latex bead-based phagocytosis assays
We exposed experimentally-treated BV-2 cells to 0.8 μm average diameter deep blue dyed uncharged latex beads (Sigma, L1398) for 1.5 hours, after which we washed, scraped, and sonicated the cell/bead suspension. We measured the absorbance of the resulting solution on a visible light spectrophotometer at 595 nm to obtain a quantitative readout for bead phagocytosis. We also exposed BV-2 cells experimentally treated as indicated for 6 hours with 1 μm average diameter fluorescent, carboxylated latex beads (Invitrogen, F8816) for 45 minutes. We washed and fixed the cells as in our ICC protocol to obtain a qualitative readout.
Microfluidic chamber system
We fabricated photoresist molds and polydimethylsiloxane (PDMS) microfluidic chambers as described by Park et al.. We plasma-cleaned 22 mm square #1.5 coverglass (Corning, Tewksbury, MA, USA, 2870–22) and submerged it in 50 μg/ml poly-d-lysine at 37°C overnight. We thoroughly washed and dried the glass and assembled the chambers by placing the molded PDMS on top of the glass and allowing it to seal. We then filled the chamber with media in the order described by Park. After removing the media, we added 20 μl of a 3.5 × 106 hippocampal cells/ml cell suspension to the upper left well of the chamber. We allowed the cells to attach for 15 minutes, and then filled the chamber with media as described by Park. We observed axons crossing the 400 μm barrier by 3 days in vitro (DIV), with extensive neuritic networks forming by 7 DIV.
All of the chambers we used for the microfluidic experiments were 7 DIV. We imaged the chambers before treatment to obtain baseline axonal images and measurements. We then plated the experimentally treated BV-2 cells in the axon compartment by carefully removing the media from the axon wells and adding 20 μl of a 3 × 106 BV-2 cells/ml cell suspension to one of the axon compartment wells. We allowed the cells to attach and then filled the chamber with media. We utilized a media pressure gradient to assure that none of the soluble experimental manipulations diffused to the cell body compartment. We cultured the chambers for a further 14 hours and then obtained post treatment brightfield images. We then deconstructed the chambers and fixed the cells with 4% paraformaldehyde (Sigma, 441244), 4% sucrose (EM Science, Gibbstown, NJ, USA, 8510) in 1x PBS for future immunocytochemistry. Based on pilot experiments with MAP-2 and Tau5 (see below), by 7 DIV, no MAP-2 positive processes were identified that traversed the entire 400 μm barrier (data not shown), suggesting that only axonal processes and growth cones were present in compartment 2.
Imaging and analysis
For our fluorescent fixed cell ICC, we imaged slides on an Olympus BX-51 upright microscope equipped with Qioptic Optigrid optical sectioning hardware (Princeton, NJ, USA). We imaged the fixed cells with a 40x 0.8NA air objective with a 0.2 um z-step or using a 20x 0.58NA air objective with a 0.5 um z-step. We captured images using a Hamamatsu ORCA-ER camera controlled by Perkin Elmer Volocity software, and created the representative images of the three-dimensional stacks using a z-projection in the same software. For our live cell bright-field images, we imaged the microfluidic chambers on an Olympus IX70 inverted microscope under 10x magnification. We captured the images using a PCO.edge camera controlled by the NIH image capturing software Micromanager
. We processed the images using NIH ImageJ
 image analysis software, with the addition of the NeuronJ Java plugin for axon length analysis (