Adult male Sprague Dawley rats (Charles River Laboratories) weighing 300-400 g were housed individually and maintained on a 12 hr light / dark cycle. Food and water were provided ad libitum throughout the experiment. Animal use procedures were in accordance with the National Institutes of Health Guide for the Care and Use of laboratory animals and were approved by the University of Texas at Dallas Animal Care and Use Committee.
Large scale AAV production/purification
An AAV2 genome plasmid harboring an RFP gene controlled by a mouse αCaMKII promoter (AAV-RFP) was packaged into functional viruses that were pseudotyped with different AAV capsid proteins to produce AAV serotypes 2/1, 2/2, 2/5, 2/7, 2/8, 2/9, 2/rh10, 2/DJ and 2/DJ8. Pseudotyped viruses were produced using a triple-transfection, helper-free method, and the resultant viruses were purified on an iodixanol step gradient, as previously described [43, 52]. For producing each virus, AAV-293 cells (Agilent Technologies) were seeded in 5×15 cm plates 24 hours before transfection so that at the time of transfection the cells would be 80-85% confluent. At the time of transfection 135 μg of pRC (serotype plasmid), 135 μg of pHelper and 135 μg of AAV2 genome plasmid harboring a red fluorescent protein gene (dsRed-Express) controlled by a mouse αCaMKII promoter (Addgene plasmid 22908)  were used per each calcium phosphate mediated transfection. The culture media was replaced 6 hours post-transfection with pre-warmed standard growth media. At Seventy-two hours post-transfection the cells were harvested by centrifugation (1500 g for 15 min), the supernatant was discarded and the cells were re-suspended in 8 ml of freezing buffer (0.15 M NaCl, 50 mM Tris, pH 8.0) and stored at -80°C until purification. For virus purification, cells were lysed by two freeze/thaw cycles in dry ice-ethanol and 42°C water baths. Five hundred units of benzonase (Novagen) was added to the lysate and incubated for 30 minutes at 37°C. The lysate was clarified by centrifugation at 3700 g for 20 min and the supernatant was applied to an iodixanol gradient column (15%, 25%, 40% and 57% iodixanol). The columns were centrifuged at 50,000 rpm for 3 hr 20 min at 10°C in a Type 70 Ti rotor (Beckman Coulter). After centrifugation the 40% iodixanol layer was extracted and transferred to a conical tube and diluted with 25 ml of PBS-MK (1×PBS without calcium or magnesium, 1 mM MgCl2, 2.5 mM KCl) and concentrated using Amicon Ultra-15 Centrifugal Filter Units (Millipore) to ~ 1 ml. This viral concentrate was diluted again with 10 ml of PBS-MK and concentrated to a final volume of 100 μl and stored at 4°C until use. All plasmids used were purified using Qiagen Endofree plasmid purification kits following the manufacturer’s instructions. AAV packaging plasmids (pRC) for different serotypes were obtained from: AAV1, AAV5 (Nicholas Muzyczka, University of Florida), AAV-2 (Agilent Technologies), AAV2*, AAVDJ and AAVDJ8 (Cell BioLabs, Inc), AAV7, AAV8, AAV9, AAVrh10 (Penn Vector Core, University of Pennsylvania). AAV2 and AAV2* are the same serotype but the AAV2 packaging plasmid was obtained from a different source (above). AAV Helper plasmid (pHelper) was obtained from (Agilent Technologies). Since calcium phosphate transfections are sensitive to the pH of the 2×HBS solution, we prepared 2×HBS at different pH (pH 7.04, 7.06, 7.08, 7.10 and 7.12) and tested them in small scale transfections in a 6 well plate with a GFP reporter plasmid to determine the optimal 2×HBS solution to use for transfection. Transfections to produce AAV serotypes 1 and 5 were performed as described above with the exception that 270 μg of the AAV1 and AAV5 serotypes plasmids were transfected and the pHelper plasmid was not transfected since the packaging plasmid for these serotypes contained both capsid sequences and helper sequences required for making AAV viruses. AAV-misc is essentially the same AAV shRNA genome plasmid that has been previously described  with the exception that the U6 based shRNA expression cassette has been replaced with a H1 based shRNA expression cassette. AAV-misc based viruses were produced and purified as described above.
Purified AAV viruses were titered using a quantitative-PCR based titering method. Five μl of purified viruses were treated with 20 Units of DNase I (Roche) at 37°C for 30 min and then further diluted in (10 mM Tris pH 7.4, 10 μg/ml of yeast tRNA solution (Ambion)) to make the final concentration of virus a 1:10,000 dilution with respect to the undiluted virus. Five μl of the diluted virus was used per a standard 20 μL Taqman PCR assay (Applied Biosystems). One μl of a 20X Taqman custom RFP Primer/Probe for AAV-RFP (RFP FP: AGCGCGTGATGAACTTCGA,
RFP RP: GCCGATGAACTTCACCTTGTAGAT, RFP Probe: 6FAM-ACCCAGGACTCCTCC) (as previously described ) or GFP Primer/Probe (ID# Mr04329676_mr, Invitrogen) for AAV-misc was used per reaction. Samples were prepared and loaded onto a 96 well plate in triplicate and quantitated using a CFX96 Real-time PCR system (BioRad) using the standard cycling parameters specified by Applied Biosystems. AAV-RFP and AAV-misc viral genome copies were quantitated based on a standard curve prepared by serially diluting the pAAV-αCaMKII-RFP and pAAV-misc plasmids respectively in (10 mM Tris pH 7.4, 10 μg/ml of yeast tRNA solution) across eight, three fold serial dilutions ranging from 1 × 103 to 3 × 106 copies of the viral genome plasmid per PCR reaction (as described above). Final viral titers were computed based on the standard curve and reported as genome copies GC/ml.
Viral titering of small scale viral crude lysate
Small scale transfections for AAV-RFP serotypes were carried out in 12 well plates using standard Lipofectamine transfection protocol (Life Technologies). Twenty four hr before transfection, AAV-293 cells were seeded in 12 well plates in AAV-293 growth media without antibiotics so that on the day of transfection the cells were 90-95% confluent. At the time of transfection, .53 μg of pRC (serotype plasmid), .53 μg of pHelper and .53 μg AAV2 genome plasmid were transfected into each well following the manufacturer’s instructions (Invitrogen). Seventy two hours post transfection the cells were harvested by centrifugation at 1500 g for 15 min and the cell pellet was re-suspended in 36.5 μl of freezing buffer. The cells were then lysed by two freeze/thaw cycles in a dry ice-ethanol bath and a 42°C water bath. Two and one half units of benzonase (Novagen) were added to the lysate and incubated for 30 min at 37°C. The lysate was then clarified by centrifugation at 3700 g for 20 min and the supernatant was used for PCR based viral titering. The partially purified viruses were treated with 20 Units of DNase I (Roche) at 37°C for 30 min and then diluted 1:1000 in (10 mM Tris pH 7.4, 10 μg/ml of yeast tRNA solution) and were titered similarly as the large scale AAV virus titering. The titers were reported as GC/ml. For this experiment all samples were processed together in triplicate to eliminate inter-experiment variability.
Basolateral complex (BLA) viral infusions
Thirty-three gauge custom made infusion cannula (C315G, Plastics One) were inserted into ~20 inch long polyethylene tubing (I.D. 0.015 in, O.D. 0.043 in, wall thickness 0.0140 in) (A-M systems, Inc.). These tubes were backfilled with sesame oil and then attached to 2 μl, 23-gauge (88500) stainless steel Hamilton syringes (Hamilton Company). Under a mixture of Ketamine (100 mg/kg) and Xylazine (10.0 mg/kg) anesthesia, rats were stereotaxically implanted bilaterally with a 33 gauge infusion cannula (described above) targeting the BLA [AP -2.9, ML ±5.2, DV -8.6]. For experiments depicted in Figures 4 and 5, 1.4 μl of the highest titer for each serotype (see Figure 2C), were bilaterally infused for 15 minutes at a rate of 0.09 μl/min. For experiments depicted in Figures 6 and 7, the viral titers for each serotype were adjusted to 7.8E + 11 GC/ml using PBS-MK and 1 μl was bilaterally infused for 15 min at the rate of 0.07 μl/min. Following infusions, the infusers were left in for 10 additional min to allow for diffusion of the virus away from the cannula after which they were withdrawn and the incision was closed using 9 mm wound clips (Mikron Precision, Inc.). The rats were allowed one week for recovery and the wound clips were removed using a wound clip remover (Mikron Precision, Inc.). Three weeks post infusion, the rats were anesthetized with an overdose of chloral hydrate (250 mg/kg) and then perfused with phosphate-buffered saline (1× Phosphate buffer, 150 mM NaCl) and 10% buffered Formalin (Fisher Scientific). The brains were fixed in 10% formalin for 4-5 hours followed by cryoprotection in 1×PBS pH7.4, 30% sucrose for 4-6 days.
Imaging and quantification of viral transduction
Following cryoprotection, the brains were frozen and 40 μm coronal sections were obtained using a cryostat which included the entire amygdala (Bregma -4.16 to -1.80). Every alternate section was placed on superfrost slides (Fisherbrand) typically yielding ~40 sections per amygdala. The remaining slices were stored in a solution of PBS-MK with 0.1 mM sodium azide. The sections on superfrost slides were imaged using an Olympus IX51 inverted fluorescent microscope under RFP and bright field filters and the images were acquired using an Olympus DP72 Digital Camera and Cellsens software (Olympus). Approximately 2500 images were taken and analyzed for this project. For experiments depicted in Figures 4 and 5, all fluorescence images were taken under the same exposure conditions of 67 ms at 20X magnification. For experiments depicted in Figures 6 and 7, all fluorescence images were taken under the same exposure conditions of 1 s at 20X magnification. As a result, images between these experiments cannot be compared directly. All ~40 sections per amygdala were quantified for viral transduction efficiency using Image-J software to specifically measure spread of transduction (Area (μm2)), mean red fluorescence intensity per transduced region (optical density (OD)), and mean viral transduction (Area*OD). For experiments depicted in Figures 6 and 7, in addition to these three measurements, the mean number of cells transduced was also determined. Because targeted viral infusions within the rodent brain even under the best circumstances will not entirely localize viral transduction to the intended target region, all 4 of these measures were taken for the total viral transduction within each imaged coronal brain slice (designated as Total within the figures), and the viral transduction which occurred only within the borders of the BLA (designated as BLA only within the figures). The total viral transduction (Area*OD) per slice was calculated by subtracting the background mean optical density (OD) from the mean OD of the total area transduced in each slice to create the OD of transduction per slice. This value was multiplied by the total area transduced within each slice, measured in μm2 to yield a measure of total viral transduction per slice. The total viral transduction per amygdala was recorded as the sum of all measures of total viral transduction per slice per amygdala and the total viral transduction per serotype was reported as the mean total viral transduction of all amygdala infusions per serotype demonstrating transduction. OD of total transduction was calculated for every slice and averaged across all slices per amygdala to yield total OD of transduction per amygdala and this value was averaged among all infusions of a particular serotype demonstrating transduction to yield mean total OD of viral transduction per serotype. The total spread of transduction per amygdala was recorded as the sum of total area transduced in all slices per transduced amygdala and this value was averaged among all infusions of a particular serotype demonstrating transduction to yield mean total viral spread per serotype. For BLA only viral transduction measurements, the OD and viral spread were calculated as above with the exception that the measurements were limited to the boundaries of the BLA in each slice and these measurements (BLA only) are reported side by side with the Total measurements within the figures. In experiments depicted in Figures 4 and 5, approximately 40 slices per amygdala were analyzed and a total of 4 amygdala per serotype were analyzed, with the exception of AAV-DJ which 5 amygdala were analyzed. A total of 2-3 rats per serotype were analyzed. For experiments depicted in Figures 6 and 7, in addition to the above mentioned measurements we also counted the number of transduced cells. To quantify the number of cells transduced, images were imported into ImageJ and converted to grayscale 8 bit images and the background subtraction function was applied. Next the threshold was set to 27 for the lower level and 255(max) for the higher level. The watershed function was applied and the cells within each image were counted for the entire image and cells only within the borders of the BLA were also counted (BLA only). The cell counts were summed across all sections that exhibited viral transduction per amygdala and cell counts per amygdala per serotype were averaged to give mean cell counts per serotype. For experiments depicted in Figures 6 and 7, approximately 40 slices per amygdala were analyzed and a total of 7-10 amygdala per serotype across 4-8 rats per serotype were analyzed (AAV2/1, AAV2/7, n = 7; AAV2/5, AAV2/8, AAV2/9, AAV2/rh10, n = 8; AAV2/DJ, n = 9; AAV2/DJ8, n = 10; n refers to number of amygdala per serotype analyzed). Non-parametric statistics were applied to evaluate statistical differences in transduction efficiency among serotypes examined using the Kruskal-Wallis test with a Dunn-Bonferroni post hoc test. Differences were considered significant if, p < 0.05 (uncorrected for multiple comparisons); however in many cases, significance was reached when correcting for multiple comparisons. Tables containing the p-values from statistical comparisons among the serotypes for mean viral transduction (total and BLA only), mean viral spread (total and BLA only) and mean number of transduced cells (total and BLA only) are provided in (Additional file 1: Table S1).
To generate the phylogram comparing the VP1 capsid proteins for the AAV serotypes used in this study, the VP1 amino acid sequences for all of the serotypes were first aligned using ClustalOmega (EMBL-EBI)  using the default parameters. The resulting phylogenetic tree output file was then imported into Phylodendron software (version 0.8d, beta January 1999; http://iubio.bio.indiana.edu/treeapp/) to generate the phenogram. To produce the sequence similarity/divergence table for the AAV serotype VP1 capsid proteins, the VP1 amino acid sequences for all serotypes were imported into MegAlign Software version 3.05a (DNASTAR, Inc.) and aligned using the Clustal method and the pair-wise sequence distances were computed using the Clustal method with a PAM250 residue weight table.
For immunohistochemistry (IHC), 1 μl of virus was bilaterally infused into the BLA at a titer of 7.8E + 11 GC/ml as described above. Twenty one days post infusion, the rats were anesthetized with an overdose of chloral hydrate (250 mg/kg) and then perfused with 4% Paraformaldehyde in 1 × PBS (pH = 7.4). Following brain extraction, the brains were fixed in 4% Paraformaldehyde in 1×PBS (pH = 7.4) for 4-5 hr, followed by cryoprotection in 1×PBS pH7.4, 30% sucrose for 4-6 days. Following cryoprotection, the brains were frozen and 40 μm coronal sections were obtained using a cryostat as described above. For NeuN IHC, brain slices were rinsed in 1×PBS for 10 min followed by incubation for 30 min with blocking buffer (1× PBS, 3% normal donkey serum, 0.3% Triton X-100). Next the slices were incubated overnight at 4°C with NeuN Antibody (1:500; MAB377 Millipore, Billerica, MA), diluted in blocking buffer. For the αCaMKII IHC, brain slices were rinsed in 1×PBS for 10 min followed by incubation for 60 min with blocking buffer (1X PBS, 5% normal donkey serum, 0.3% Triton X-100). Next the slices were incubated overnight at 4°C with αCaMKII Antibody (1:300; 05-532, Millipore), diluted in antibody dilution buffer (1XPBS, 1% BSA, 0.3% Triton X-100). For secondary antibody staining 1:500 dilution of fluorescein isothiocyanate (FITC) conjugated anti-Mouse IgG (AP192F; Millipore) was used and the slices were incubated for 2 hr at room temperature. The slides were coverslipped with Vectashield HardSet Mounting Medium (Vector Laboratories). IHC images were captured at 100× magnification using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software. The number of AAV transduced (RFP) cells within the BLA that were also NeuN positive or αCaMKII positive was determined by quantifying the number of RFP cells that also co-localized with the NeuN or αCaMKII FITC signal within an area of 135173.56 μm2 from within the BLA of each slice examined. The area within each slice quantified typically contained ~ 65 RFP positive cells and a total of three slices were counted per serotype. Cells were manually quantified using Adobe Photoshop CS5 software and the colocation results were reported as a percentage of total RFP cells counted. Variability in co-localization results between slices was reported as standard error of the mean.