The production of viral vectors designed to express large and difficult to express transgenes within neurons
© Holehonnur et al.; licensee BioMed Central. 2015
Received: 9 December 2014
Accepted: 2 February 2015
Published: 24 February 2015
Viral vectors are frequently used to deliver and direct expression of transgenes in a spatially and temporally restricted manner within the nervous system of numerous model organisms. Despite the common use of viral vectors to direct ectopic expression of transgenes within the nervous system, creating high titer viral vectors that are capable of expressing very large transgenes or difficult to express transgenes imposes unique challenges. Here we describe the development of adeno-associated viruses (AAV) and lentiviruses designed to express the large and difficult to express GluN2A or GluN2B subunits of the N-methyl-D-aspartate receptor (NMDA) receptor, specifically within neurons.
We created a number of custom designed AAV and lentiviral vectors that were optimized for large transgenes, by minimizing DNA sequences that were not essential, utilizing short promoter sequences of 8 widely used promoters (RSV, EFS, TRE3G, 0.4αCaMKII, 1.3αCaMKII, 0.5Synapsin, 1.1Synapsin and CMV) and utilizing a very short (~75 bps) 3′ untranslated sequence. Not surprisingly these promoters differed in their ability to express the GluN2 subunits, however surprisingly we found that the neuron specific synapsin and αCaMKII, promoters were incapable of conferring detectable expression of full length GluN2 subunits and detectable expression could only be achieved from these promoters if the transgene included an intron or if the GluN2 subunit transgenes were truncated to only include the coding regions of the GluN2 transmembrane domains.
We determined that viral packaging limit, transgene promoter and the presence of an intron within the transgene were all important factors that contributed to being able to successfully develop viral vectors designed to deliver and express GluN2 transgenes in a neuron specific manner. Because these vectors have been optimized to accommodate large open reading frames and in some cases contain an intron to facilitate expression of difficult to express transgenes, these viral vectors likely could be useful for delivering and expressing many large or difficult to express transgenes in a neuron specific manner.
Viral vectors have become a common tool in neuroscience to express transgenes in a spatially and temporally restricted manner within the nervous system of numerous model organisms [1-5]. Among the various viral vector systems that are currently available, the most commonly used systems for targeting the nervous system are adeno-associated viruses (AAV) and lentiviruses.
Adeno-associated viruses are single stranded DNA viruses. Numerous serotypes of AAV possess a natural tropism for neurons and when these serotypes of AAV are infused into the CNS, neurons are the predominant cell type transduced. This property of AAV makes it convenient for directing ectopic transgene expression to neurons and eliminates the need to use neuronal specific promoters (i.e. synapsin, αCaMKII) to confine transgene expression to neurons [6-8]. Alternatively, lentiviruses derived from human immunodeficiency virus type 1 (HIV-1), are single stranded RNA viruses that are commonly pseudotyped with the vesicular stomatitis virus G-protein (VSV-G) and the VSV-G capsid provides the virus the ability to transduce many cell types. However due to this naturally broad tropism, VSV-G pseudotyped lentiviruses transduce glia cells very efficiently [9-12]. Therefore if neuronal specific gene expression is desired, then transgenes within VSV-G pseudotyped viruses must be controlled by neuronal specific promoters.
Despite the ubiquitous use of viral vectors to direct ectopic expression of trangenes within the nervous system, creating high titer viral vectors that are capable of expressing very large transgenes or difficult to express transgenes imposes unique challenges and this is in part due to limitations in the maximum viral genomes sizes that can be packaged into functional viruses. For example, AAV has an optimal viral genome size limit of ~4.7 - 5 Kb, which includes the ~150 bp inverted terminal repeats (ITRs) present at each end of the viral genome (combined ITR length is ~300 bps). This limit of course restricts the amount of genetic information that can be carried by the virus and therefore restricts the size and/or number of genes that can be packaged into AAV. Withstanding these limitations, there is convincing evidence that viral genomes as large as ~5.2 Kb can be packaged into functional viruses and even larger genomes have been used to create viruses that can confer ectopic transgene expression. It is thought that genomes larger than ~5.2 kb become truncated to ~5.2 kb during viral packaging and recombination between truncated genomes in the host cell enables the reconstitution of a functional transgene . Producing AAV with genomes that are larger than the optimal genome size limit generally leads to a reduction in viral titer and a reduction of transgene expression. For in vitro purposes, this may be tolerable since usually a larger volume of virus can easily be used . However for in vivo purposes the amount of virus that can be infused is constrained significantly. For example infusions of virus into the brain of a rodent can usually be no larger than a few microliters. It currently remains unclear how practical it is to use AAV with larger genomes for in vivo use. Lentiviruses possess a higher packaging limit than does AAV (~10 kb), but lentiviruses also exhibit reduced titers as viral genome size increases .
We were interested in developing viruses that could deliver a transgene to brain cells that would express the GluN2A or GluN2B subunit of the NMDA receptor, specifically within neurons. NMDA receptors are ionotropic glutamate receptors that are thought to be involved in synaptic plasticity, and multiple forms of learning  and their dysregulation are thought to occur in numerous psychiatric disorders [17,18]. Studies indicate that NMDAR signaling can be greatly influenced by the GluN2A/GluN2B ratio and this is due to inherent differences in calcium influx and cellular signaling mediated by these subunits . Notably a switch in the GluN2A/B ratio occurs during development [19-21], in the aging brain [22-25], following synaptic activity [26-29] and irregularities of this ratio occur in models of psychiatric disease . Therefore it would be of value if specific GluN2 subunits could be over expressed in models of learning and psychiatric disease, especially within specific circuits of the brain or at specific time points during development or before or after behavioral training to alter the GluN2A/B ratio. One way to accomplish this is by using viral vectors that could deliver these transgenes to neurons within the brain. However, the GluN2A and GluN2B coding regions are 4392 and 4446 bps respectively. Their large size creates challenges for creating a viral vector that can be produced at high titers and capable of conferring the expression of these transgenes efficiently in vivo in a neuron specific manner. Therefore to accomplish this goal, we systematically created a number of custom designed AAV and lentiviral vectors that were optimized for large transgenes, by minimizing DNA sequences that were not essential, utilizing short promoter sequences of 8 widely used promoters RSV, EFS, TRE3G, 0.4αCaMKII, 1.3αCaMKII, 0.5Synapsin, 1.1Synapsin and CMV, and utilizing a very short (~75 bps) 3′ untranslated sequence containing an SV40 based poly adenylation signal. Not surprisingly these promoters differed in their ability to confer expression of the GluN2 subunits, however surprisingly we found that the synapsin and αCaMKII, neuron specific promoters were incapable of conferring detectable expression of full length GluN2 subunits and detectable expression could only be achieved if the transgene included an intron or if the GluN2 subunit transgenes were truncated to only include the coding regions of their transmembrane domains. Collectively this study determined that the GluN2 subunits are difficult to express and that viral packaging limit and transgene promoter choice were critical factors for developing viral vectors designed to express GluN2 subunit transgenes. We suspect that many of the custom viral vectors we created could be of use by other investigators since they are optimized for larger transgenes. Additionally we created a lentiviral vector designed to express difficult to express transgenes specifically within neurons – this vector contains an intron downstream of a 1.3αCaMKII promoter. Here we describe our findings.
All viral vector plasmids were created using standard recombinant cloning procedures. A detailed description of how each viral plasmid was created for this study is described in Additional file 1. Viral genome sizes, and DNA oligonucleotides used for cloning purposes are provided in Additional file 2. Promoter DNA sequences are provided in Additional file 3. Additional relevant DNA sequences are provided in Additional file 4. DNA vector sequences are provided for all viral vectors created for this study in Additional file 5. AAV2 genome vectors plasmids, pAAV-Basic (Vector Biolabs) and pAAV (previously described, generous gift from Ralph DiLeone  were used for this study. The lentiviral plasmid pLenti7.3 DEST (Invitrogen) was modified and used for all lentiviral vectors for this study. The Flag-tagged rat GluN2A(pRK5) and mouse GluN2B(pRK5) containing plasmids were generously provided by Katherine Roche . The 0.4αCamKII promoter was PCR amplified from rat genomic DNA. The mouse 1.3αCamKII promoter was PCR amplified from pAAV-CaMKII-RFP (Addgene # 22908) . The rat synapsin promoters were PCR amplified from the L26 plasmid generously provided by Pavel Osten . The RSV and CMV promoters were PCR amplified from pLenti7.3 DEST (Invitrogen). The TRE3G and EFS promoters were PCR amplified from Addgene vectors #(27569) and #(49535)  respectively. The AAV plasmids were propagated at 37°C in One shot Top Ten competent cells (Invitrogen). The lentiviral plasmids were propagated at room temperature in Stbl2 competent cells (Life Technologies). DNA plasmids were verified by DNA sequencing to be correct. Viral vector plasmids will be provided upon request and many of these will be made available through Addgene.
293FT (Invitrogen) and Neuro2A (N2A) (ATCC, Cat #CCL-131) cells were grown in Dulbecco’s Modified Eagle Medium supplemented with 5% Glucose, 5 mM sodium pyruvate and 10% Fetal Bovine Serum (Life Technologies) in 5% CO2 humidified cell culture incubator.
293FT and N2A cells were seeded at 65% confluency on pre-coated poly-l-lysine (0.1 mg/mL, Sigma) coverslips in 24-well plates. On the day of transfection, 0.8 ug of plasmid DNA was transfected into each well using the standard lipofectamine protocol (Life technologies). Plasmids containing EFS, RSV, TRE3G, CMV promoters were transfected into 293FT cells and plasmids expressing αCaMKII and synapsin promoters were transfected into N2A cells. After 24 hours, the coverslips were fixed using 4% paraformaldehyde in 1XPBS (pH7.4) (Fisher Scientific) for 20 minutes followed by a wash with 1X PBS(pH7.4) and then incubated with 0.2% Triton-X in 1XPBS (pH7.4) (American Bioanalytical) for 10 minutes on ice. The coverslips were next blocked in blocking serum (5% BSA (Sigma), 5% Goat serum (Sigma) in 1XPBS (pH7.4) for 90 minutes at room temperature. Following this, coverslips were incubated in 40 μls of mouse monoclonal anti-Flag primary antibody in blocking serum (1:200;F1804, Sigma) for 2 hours at 37°C. After the primary antibody incubation, the coverslips were washed twice with 1XPBS (pH7.4) and incubated with 40 μls of anti-mouse Texas Red antibody in blocking serum (1:1000, Life technologies) for 1 hour at room temperature. The coverslips were then rinsed thrice with 1XPBS (pH7.4) followed by incubation with 500 μls DAPI in 1XPBS (pH7.4) (1:25,000, Invitrogen) for 10 minutes at room temperature. The coverslips were rinsed three more times with 1XPBS (pH7.4) before mounting on superfrost slides (Fisherbrand) using Fluoromount-G (Southern Biotech). All images were visualized using 200X magnification using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software. ICCs were performed in duplicates and the results were reproduced in at least 2 separate experiments. In order to compare the strength of promoters in AAV plasmid vectors, pRK5-GluN2A (positive control), AAV-CMV/TRE3G/EFS/RSV-GluN2A vectors, pTet-Off vector (Clontech) and AAV-CMV-GFP were transfected at 266.7 ngs each into one 24 well using the standard lipofectamine transfection protocol. The AAV-CMV-GFP plasmid was used to control for transfection efficiency. These transfections were performed in duplicate and were processed for ICC as mentioned above. Following this, images were acquired at 100X magnification at constant exposure time (Texas Red - 1.1 s; GFP – 66 ms) using an Olympus DP72 Digital Camera and Cellsens software (Olympus). At least 6 distinct fields of view were imaged and quantified for each group. Using Image-J software, the mean OD for Texas Red (GluN2A), GFP and DAPI were calculated. The GFP signal served as a transfection efficiency control and the DAPI signal served as a cell number control. No difference in GFP signal or DAPI signals were found among the groups indicated that the transfection efficiency and cell number did not vary significantly among these groups. To obtain the Texas Red (GluN2A) expression levels for each group, the Texas Red signal was normalized to the GFP signal and these data were plotted as average percent expression as compared to the control group. Error bars represent standard error of mean. Parametric statistics using one-way ANOVA with a Fisher PLSD post-hoc test was applied to evaluate statistical differences among AAV vector transgene expression. Differences were considered significant if, p < 0.05.
AAV virus production/purification
An AAV2 genome plasmid harboring the gene of interest (Flag-GluN2A, Flag-GluN2B, GFP) was packaged into functional viruses that were pseudotyped with AAV DJ or AAV DJ8 capsid proteins to produce AAV2/DJ and AAV2/DJ8 viruses. Pseudotyped viruses were produced using a triple-transfection, helper-free method using 293FT cells and the resultant viruses were purified on an iodixanol step gradient, as previously described [31,35]. Purified viruses were titered using a qRT-PCR based method to determine the number of DNAse resistant viral particles and the results were reported as genomes copies (GC)/mL and this was performed as previously described  using PCR primers specific for GluN2A (Life Technologies, Cat#Rn00561341_m1), GluN2B (Life Technologies, Cat# Mm00433820_m1) and GFP (Life technologies, Cat# Mr04329676_mr).
Lenti virus production/purification
Small scale lentiviruses were made following the Invitrogen Forward Transfection protocol. Briefly, the cells were seeded at 65% confluency in pre-coated poly-l-lysine (0.1 mg/mL, Sigma) 10 cm plates a day before transfection. On the day of transfection, 3 μg of the lentiviral genome plasmid containing the gene of interest and 9 ug of the lentiviral packaging plasmids: pLP1 (3.34 μg), pLP2 (2.21 μg) and pLP/VSV-G (3.34 μg) were transfected using the Virapower Lentivirus Forward Transfection (Life Technologies) protocol per 10 cm plate. At seventy-two hours post transfection, the supernatant was collected and centrifuged at 3000 g for 15 minutes and filtered through 0.45 μm PVDF filters (Fisher brand) to remove cell debris. Following this, the lentivirus was concentrated using PEG-it (System Biosciences). At Forty-eight hours following addition of PEG-it, the viral solution was centrifuged using a swinging bucket rotor at 1500 g for 30 minutes to pellet the virus and the viral pellet was resuspended in 100 μls of cold 1XPBS (pH7.4). The lentiviruses were then aliquoted in 10 μl volumes and frozen at −80°C. Large scale lentivirus production was performed as previously described [36,37]. Briefly, 3 (15 cm) plates pre-coated with poly-l-lysine (0.1 mg/mL, Sigma) were transfected with 27 ug of lentiviral genome plasmid containing the gene of interest and 84 ug of lentiviral packaging plasmids pLP1 (31.5 μg), pLP2 (21 μg) and pLP/VSV-G (31.5 μg) diluted in 10.5 ml of optimem and mixed with 240 μl of lipofectamine diluted in 10.5 ml optimem (5 minute incubation) and was further incubated for 20 minutes. Following this, existing media in each plate was replaced with fresh 13 ml of optimem (Life Technologies) and 7 ml of the transfection mixture was added. At twenty hours post transfection, the existing media from each plate was replaced with 22 ml of warm optimem media. Forty eight hours after transfection, the supernatant was collected, centrifuged for 1200 rpm for 15 minutes at 4°C followed by filtering through 0.45 μm PVDF filter (Millipore). The viral supernatant was then added into Beckmann ultracentrifuge tubes (344058) and spun at 26,000 rpm for 90 minutes at 4°C in a Beckmann SW 32-Ti swinging bucket rotor. Following ultracentrifugation, the supernatant was discarded carefully and the viral pellet in each tube was resuspended in 40 μl of cold 1XPBS. The viruses were then aliquoted into 10 μl aliquots and stored at −80°C. Purified lentiviruses were titered using an qRT-PCR based method (adapted from Global UltraRapid Lentiviral Titer Kit, System Biosciences) to determine the number of integrated lentiviral genomes directly from lysates of cells and the results were reported as genomes copies (GC)/mL. Briefly, 2 μl of concentrated lentivirus were added to 293FT cells grown in 24 well plates, in triplicate. Lentiviruses concentrated using ultracentrifugation appeared to transduce cells in culture poorly when compared to viruses concentrated using PEG-it reagent, therefore we used TranDux reagent (System Biosciences) to enhance in vitro transduction of ultracentrifugation concentrated lentiviruses. At seventy-two hours after addition of the viruses, the cells were harvested using 75 μl of lysis buffer (System Biosciences) and heated for 2 minutes at 95°C. The supernatant was then centrifuged at 14,000 rpm for 2 minutes and 5 μl of this solution was used per standard 20 μL Taqman PCR assay (Applied Biosystems). One μl of a 20X Taqman custom WPRE Primer/Probe for pLenti-WPRE (WPRE FP: CCGTTGTCAGGCAACGTG, WPRE RP: AGCTGACAGGTGGTGGCAAT, WPRE Probe: 5′FAM-TGCTGACGCAACCCCCACTGGT-TAMRA 3′) 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 Global UltraRapid Lentiviral Titer Kit (System Biosciences). Lentiviral genome copies were quantitated based on a standard curve prepared by serially diluting the pLenti-0.4αCaMKII-GFP plasmid in (10 mM Tris pH 7.4, 10 μg/ml of yeast tRNA solution) across eight, three fold serial dilutions ranging from 1 × 105 to 1× 109 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.
AAV and lentivirus transduction in vitro
293FT cells grown in 24 well plates were each transduced with 0.8 - 1.3 μl of AAV containing ~1E + 13 GC/mL virus in 200 μl of L-DMEM (growth media with 2% FBS) media for 2 hours followed by addition of 400 μl of H-DMEM (growth media with 18% FBS) media. 293FT cells grown in 24 wells were transduced with 2, 10 or 20 μl of PEG-it purified lentivirus containing 1E + 9 GC/mL - 1E + 10 GC/mL virus diluted into 500 μl of DMEM growth media. The cells were incubated for 48 hours before being processed for ICCs as described above.
Adult male Sprague Dawley rats (Harlan) weighing 300 – 400 g and adult C57BL/6 αCaMKII-tTA mice (Jackson Laboratories, cat # 003010 B6)  weighing 20 – 25 g were housed individually and maintained on a 12 hour 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.
AAV and lentivirus transduction in vivo
Under a mixture of ketamine (100 mg/kg) and xylazine (10.0 mg/kg) anesthesia, rats and mice were stereotaxically implanted bilaterally with a 28 gauge infusion cannula targeting the basolateral amygdala (BLA) [rat: AP −2.9, ML ±5.2, DV −8.6, mice: AP −1.6 ML ±3.3 DV −4.95]. AAV and lentiviruses were infused essentially as described previously . One microliter of AAV virus at ~1E + 13 GC/mL and lentivirus at 1E + 9 GC/mL - 1E + 11 GC/mL were bilaterally infused into the rat BLA over 15 minutes at a rate of 0.07 μl/min. Half of one microliter of lentiviruses ~1E + 11 GC/mL designed to express TRE3G-GFP and TRE3G-Flag-GluN2A/B were bilaterally infused into the BLA of αCaMKII-tTA mice over 15 minutes at the rate of 0.07 μl/min.
Twenty one days post infusion of AAV viruses and ten days post infusion of lentiviruses, the animals were anesthetized with an overdose of chloral hydrate (250 mg/kg) and then perfused with 1XPBS-A (1X phosphate buffer, 150 mM NaCl) and 10% buffered formalin (Fisher scientific) for animals infused with GFP viruses or with 1XPBS (pH = 7.4) (American Bioanalytical) and 4% paraformaldehyde in 1XPBS (pH7.4) for animals infused with GluN2 viruses. Following brain extraction, the brains were fixed in formalin/4% paraformaldehyde (Fisher scientific) in 1XPBS (pH7.4) for 3–4 hours, followed by cryoprotection in 30% sucrose 1XPBS-A (1X Phosphate buffer, 150 mM NaCl)/1X PBS pH7.4, for 3–4 days. Following cryoprotection, the brains were frozen and 40 μm coronal sections were obtained using a cryostat. For viruses designed to express GFP, images were obtained at 100X using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software. For viruses designed to express Flag-GluN2A/GluN2B, the slices were processed for Flag-IHC. The slices were first rinsed with 1XPBS (pH7.4) for 10 min followed by incubation for 60 min with blocking buffer (1X PBS, 5% normal goat serum, 0.3% Triton X-100). Next the slices were incubated overnight at 4°C with Flag antibody (1:800; M2 8146, Cell signaling), diluted in antibody dilution buffer (1XPBS, 1% BSA, 0.3% Triton X-100). For secondary antibody staining, the slices were incubated for 2 hours at room temperature in a solution of 1:1000 dilution of anti-mouse Texas Red antibody (Life Technologies). The slides were coverslipped with Vectashield Hard Set Mounting Medium (Vector Laboratories). IHC images were captured at 100X magnification using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software. For Flag and NeuN co-localization IHC, the slices were processed with Flag and NeuN antibodies. Briefly, the slices were prepared and fixed as mentioned above. The slices were first rinsed with 1X-PBS (pH7.4) for 10 min followed by incubation for 60 min with blocking buffer (1X PBS, 2.5% normal goat serum, 2.5% normal donkey serum, 0.3% Triton X-100). Next the slices were incubated overnight at 4°C with an Anti-Flag antibody (1:500; PA1-984B, Thermoscientific and Anti-NeuN antibody (1:500; MAB377 Millipore, Billerica, MA) diluted in antibody dilution buffer (1XPBS, 1% BSA, 0.3% Triton X-100). For secondary antibody staining, 1:1000 dilution of anti-rabbit Texas Red antibody (Life Technologies) and 1:200 dilution of fluorescein isothiocyanate (FITC) conjugated anti-Mouse IgG (AP192F; Millipore) was used. The slices were incubated for 2 hours at room temperature. The slides were coverslipped with Vectashield Hard Set Mounting Medium (Vector Laboratories). IHC images were captured at 100X magnification using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software under Tx-red and FITC channels. The number of transduced (Flag-staining) cells that were also NeuN positive was determined by quantifying the number of cells showing Flag-staining that also co-localized with the NeuN FITC signal within an area of 19008.78 μm2 per slice, (n = 3). Cells were manually quantified using Adobe Photoshop CS5 software and the co-localization results were reported as a percentage of total Flag-staining positive cells counted.
Neuron specific promoters are incapable of conferring detectable expression of GluN2A/B, but they can confer expression of GluN2A/B C-terminal tail deletion mutants and GFP
AAV was not capable of conferring full length GluN2A/B transgene expression in vitro or in vivo
Lentiviral plasmids confer spurious transgene expression in vitro, due to the presence of the RSV promoter at the 5′ end of the lentiviral genome
Lentiviruses designed with TRE3G and CMV promoters conferred full length GluN2 expression in vitro and in vivo
Next we wanted to determine if the lentiviral plasmids designed to express GluN2A/B and GFP could be packaged into functional virus and if these viruses could be used to transduce cells in vitro and in vivo to induce ectopic transgene expression. Lentiviruses were produced using the following lentiviral vector plasmids: 0.4αCaMKII-GluN2A, 0.4αCaMKII-GluN2B, 1.3αCaMKII-GluN2A, 1.3αCaMKII-GluN2B, 1.1Syn-GluN2A, 1.1Syn-GluN2B, TRE3G-GluN2A, TRE3G-GluN2B, CMV-GluN2A, CMV-GluN2B, 0.4αCaMKII-GFP, 1.3αCaMKII-GFP, 1.1Syn-GFP, TRE3G-GFP, and CMV-GFP. The viruses that contained transgenes controlled by the CMV and TRE3G promoters were applied to 293FT cells in vitro, and 48 hours post application, the cells were examined for transgene expression as described above.
Lentivirus designed with a 1.3 αCaMKII promoter and an intron conferred in vivo expression of full length GluN2A
Creating viral vectors that can deliver large and or difficult to express transgenes to cells to direct robust ectopic expression is not trivial, however our findings coupled with the development of specific viral vectors should benefit others attempting to achieve this. We determined that viral packaging limit, transgene promoter and the presence of an intron within the transgene were all important factors that contributed to being able to successfully develop viral vectors designed to deliver and express GluN2 transgenes in a neuron specific manner. We initially developed lentiviruses designed to express the GluN2 transgenes with synapsin and αCaMKII promoters, but these viruses were never capable of conferring transgene expression in vitro or in vivo and it was unclear why. Because we could not initially develop lentivirus for this purpose, we began to work with AAV, despite knowing that AAVs packaging limit could create technical limitations. However working with AAV turned out to be fortuitous because it was during these experiments that we discovered that the synapsin and αCaMKII promoters were unable to confer expression of the GluN2 subunits, and that the transgene expression that was observed from lentiviral plasmids was in fact due to the lentiviral plasmid RSV promoter and not due to the neuron specific promoters. These findings were surprising considering that the synapsin and αCaMKII promoters are routinely used to direct neuron specific gene expression so we did not anticipate that they would not be capable of conferring detectable GluN2 expression. We were able to create functional lentiviruses that could express full length GluN2 transgenes using CMV and TRE3G promoters. And if the lentiviral transgene included an intron between the 1.3αCaMKII promoter and the GluN2A coding region, the 1.3αCaMKII promoter was capable of conferring neuron specific expression of GluN2A and functional viruses could be produced with this transgene. We also created this lentiviral plasmid with a multiple cloning site, so that any transgene can easily be cloned into it. The 1.3αCaMKII promoter-intron-MCS vector likely would be useful for anyone attempting to develop a lentivirus designed to express a difficult to express transgene that requires neuron specific expression.
Despite our best efforts, we were not able to create a functional adeno-associated virus that could confer full length GluN2 expression. These vectors were optimized so that they could be used for genes with large coding regions and this was accomplished by removing unnecessary DNA from the viral vector and by using short promoters and a ~75 bps 3′UTR that contained an SV40 based poly adenylation signal. However even with these modifications, the AAV-GluN2 viral genomes were all above 5 kb and this was due to the fact that the GluN2 coding regions are so large (~4.4 kb). It’s not necessarily surprising that the AAV-CMV-GluN2A, AAV-CMV-GluN2B, and AAV-TRE3G-GluN2A viruses were not capable of conferring GluN2 transgene expression, since these viruses were constructed with genomes that were much larger than the optimal genome size for AAV, (5411, 5476, 5228 bps respectively). However the AAV-EFS-GluN2A, and AAV-RSV-GluN2A possess genomes that were only marginally above the optimal packaging limit (5077 and 5055 bps respectively) and these also were not able to confer GluN2 transgene expression. Considering we have created functional AAV with genomes of similar sizes before (data not shown) and others have as well , we suspect that the AAV-EFS-GluN2A and AAV-RSV-GluN2A viruses were not capable of conferring expression of the GluN2 transgenes in part due to the less than optimal genome sizes and the fact that the EFS and RSV promoters do not confer robust expression of the GluN2 transgenes. Had we been able to find a promoter that was as short as or shorter than the RSV promoter (~225 bps), but also of similar strength to the CMV promoter, we suspect we would have been able to create a functional AAV that could confer GluN2 transgene expression. Regardless, these vectors have been optimized for transgenes with large coding regions and the existing GFP/GluN2 transgenes can easily be removed and any gene of interest can easily be cloned into these vectors – we suspect these viral plasmids can be useful for investigators interested in using AAV to express large transgenes.
There has been a desire to produce viruses that can deliver large transgenes for human gene therapy purposes and consequently there have been a number of studies examining this problem. For example numerous laboratories have been working on ways to develop AAV so it can deliver the large Cystic fibrosis transmembrane conductance regulator (CFTR) gene. The coding region of CFTR is ~4.6 kb. Previous studies have determined that the use of short promoters and short 3′ UTRs and the inclusion of an intron have facilitated production of AAV that can deliver and express CFTR [39,40] and these approaches were similar to what we have used to design viruses that can deliver GluN2 transgenes. However considering the CFTR transgene is capable of being expressed from significantly truncated promoters, it seems likely that the CFTR is not as difficult to express as the GluN2 transgenes. However more research would be required to determine this. Other approaches to develop AAV, so it can deliver large transgenes has involved a dual AAV system, where one AAV contains the promoter and some of the protein coding region of the transgene of interest and the other AAV contains the remaining protein coding region and the 3′ UTR containing a poly adenylation signal. These two AAVs also contain intron splice acceptor and donor signals and bridging sequences that are appropriately placed, so when these AAVs transduce cells, and the AAV genomes concatemerize, the complete transgene can be reconstituted, where the intron bridges the two viral genomes, resulting in an intact transgene. This approach holds promise for delivering large transgenes, but this approach also would result in reduced transgene expression compared to a single AAV that contained a complete transgene that was within the optimal AAV packaging limit . Recently particular poly adenylation signal sequences have been identified that are short and therefore allow transgenes with larger coding regions to be packaged into AAV and these poly adenylation signal sequences were found to increase transgene expression . In our study we included an intron within our lentiviral vector to facilitate GluN2A expression; however others have taken an alternative approach where they created hybrid promoters, by combining neuron specific promoters with portions of the CMV enhancer . These hybrid promoters were capable of increasing expression, but in some cases these hybrid promoters exhibited less neuronal specificity.
We found that the neuron specific synapsin and αCaMKII, promoters were incapable of conferring detectable expression of full length GluN2 subunits and detectable expression could only be achieved from these promoters if the transgene included an intron or if the GluN2 subunit transgenes were truncated to only include the coding regions of the GluN2 transmembrane domains. It remains unclear if these transgenes were difficult to express simply due to their size or if there is something inherent to the mRNA regions that code for the C-terminal regions of the GluN2 protein or the C-terminal protein regions themselves that makes it difficult to express the full length GluN2 gene products. Recently AAV vectors were produced that were capable of delivering and expressing the Cas9 gene within neurons. The coding region of Cas9 is ~4.2 kb, which is slightly smaller than the coding regions for GluN2A and GluN2B. These vectors were designed with a truncated MeCP2 promoter and a short poly adenylation signal . It remains to be determined how strong of a promoter this short MeCP2 promoter is and if it can confer neuronal specificity when used within lenti viruses. Considering Cas9 is very large and it appears to be adequately expressed from this short MeCP2 promoter and an EFS promoter (Addgene # 49535) , it seems likely that Cas9 is not necessarily difficult to express. We have found that Cas9 expresses well from an EFS promoter and truncated inducible promoters (data not shown).
Our findings underscore the importance of confirming that the viral plasmid transgene functions as intended before actually producing virus. While this is straightforward when using AAV plasmids, we found that with pLenti7.3 (Invitrogen) based lentiviral plasmids, it is not straightforward, since the viral 5′ RSV promoter conferred expression of our transgenes, essentially masking our ability to detect if the actual transgene promoter was capable of conferring expression of the transgene. All lentiviral genome plasmids contain a promoter at the 5′ end of the genome just prior to the right LTR, and this is so the viral genome can be transcribed in its entirety (R-LTR - L-LTR) when producing recombinant replication deficient lentivirus in cell culture. However this promoter is not present in the packaged viral RNA genome. We have not examined if other commercially available lentiviral plasmids confer expression of the transgenes within these plasmids from the 5′ promoter, but considering that all lentiviral plasmids possess a 5′ promoter and many utilize the RSV promoter, it is likely that other lentiviral plasmids exhibit this same phenomenon and this should be taken into account when confirming if the viral transgene is functioning correctly.
In conclusion this study developed a number of viral vectors that could be of use to investigators in need of AAV and lenti vectors optimized for large or difficult to express transgenes. Additionally we have developed a number of viral vectors designed to express full length GluN2A and GluN2B, and GluN2A and B transgenes that possess deletions of their C-terminal cytoplasmic tails. These vectors could be very useful for investigators interested in examining the role of GluN2 in models of synaptic plasticity, learning and memory and psychiatric disease in vitro and in vivo. All viral vectors described in this study will be made available upon request and some of these will be made available through Addgene.
Supported by NIH grants RMH096202A and RMH100650A and the University of Texas at Dallas. We thank Katherine Roche for generously providing us with GluN2A(pRK5) and GluN2B(pRK5) plasmids. We also thank Pavel Osten and Eric Campeau generously providing us with L26 plasmid and pLenti CMVTRE3G eGFP Neo (w821-1) plasmid respectively.
- Kootstra NA, Verma IM. Gene therapy with viral vectors. Annu Rev Pharmacol Toxicol. 2003;43:413–39.View ArticlePubMedGoogle Scholar
- Lentz TB, Gray SJ, Samulski RJ. Viral vectors for gene delivery to the central nervous system. Neurobiol Dis. 2012;48(2):179–88.View ArticlePubMed CentralPubMedGoogle Scholar
- Pitkow LJ, Sharer CA, Ren X, Insel TR, Terwilliger EF, Young LJ. Facilitation of affiliation and pair-bond formation by vasopressin receptor gene transfer into the ventral forebrain of a monogamous vole. J Neurosci. 2001;21(18):7392–6.PubMedGoogle Scholar
- Carlezon Jr WA, Haile CN, Coppersmith R, Hayashi Y, Malinow R, Neve RL, et al. Distinct sites of opiate reward and aversion within the midbrain identified using a herpes simplex virus vector expressing GluR1. J Neurosci. 2000;20(5):RC62.PubMedGoogle Scholar
- Neill JC, Sarkisian MR, Wang Y, Liu Z, Yu L, Tandon P, et al. Enhanced auditory reversal learning by genetic activation of protein kinase C in small groups of rat hippocampal neurons. Brain Res Mol Brain Res. 2001;93(2):127–36.View ArticlePubMedGoogle Scholar
- Burger C, Gorbatyuk OS, Velardo MJ, Peden CS, Williams P, Zolotukhin S, et al. Recombinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. Mol Ther. 2004;10(2):302–17.View ArticlePubMedGoogle Scholar
- McFarland NR, Lee JS, Hyman BT, McLean PJ. Comparison of transduction efficiency of recombinant AAV serotypes 1, 2, 5, and 8 in the rat nigrostriatal system. J Neurochem. 2009;109(3):838–45.View ArticlePubMed CentralPubMedGoogle Scholar
- Holehonnur R, Luong JA, Chaturvedi D, Ho A, Lella SK, Hosek MP, et al. Adeno-associated viral serotypes produce differing titers and differentially transduce neurons within the rat basal and lateral amygdala. BMC Neurosci. 2014;15:28.View ArticlePubMed CentralPubMedGoogle Scholar
- Watson DJ, Kobinger GP, Passini MA, Wilson JM, Wolfe JH. Targeted transduction patterns in the mouse brain by lentivirus vectors pseudotyped with VSV, Ebola, Mokola, LCMV, or MuLV envelope proteins. Mol Ther. 2002;5(5 Pt 1):528–37.View ArticlePubMedGoogle Scholar
- Croci C, Fasano S, Superchi D, Perani L, Martellosio A, Brambilla R, et al. Cerebellar neurons and glial cells are transducible by lentiviral vectors without decrease of cerebellar functions. Dev Neurosci. 2006;28(3):216–21.View ArticlePubMedGoogle Scholar
- Bosch MK, Nerbonne JM, Ornitz DM. Dual transgene expression in murine cerebellar Purkinje neurons by viral transduction in vivo. PLoS One. 2014;9(8):e104062.View ArticlePubMed CentralPubMedGoogle Scholar
- Dittgen T, Nimmerjahn A, Komai S, Licznerski P, Waters J, Margrie TW, et al. Lentivirus-based genetic manipulations of cortical neurons and their optical and electrophysiological monitoring in vivo. Proc Natl Acad Sci U S A. 2004;101(52):18206–11.View ArticlePubMed CentralPubMedGoogle Scholar
- Wu Z, Yang H, Colosi P. Effect of genome size on AAV vector packaging. Mol Ther. 2010;18(1):80–6.View ArticlePubMed CentralPubMedGoogle Scholar
- Dong JY, Fan PD, Frizzell RA. Quantitative analysis of the packaging capacity of recombinant adeno-associated virus. Hum Gene Ther. 1996;7(17):2101–12.View ArticlePubMedGoogle Scholar
- Kumar M, Keller B, Makalou N, Sutton RE. Systematic determination of the packaging limit of lentiviral vectors. Hum Gene Ther. 2001;12(15):1893–905.View ArticlePubMedGoogle Scholar
- Miserendino MJ, Sananes CB, Melia KR, Davis M. Blocking of acquisition but not expression of conditioned fear-potentiated startle by NMDA antagonists in the amygdala. Nature. 1990;345(6277):716–8.View ArticlePubMedGoogle Scholar
- Lau CG, Zukin RS. NMDA receptor trafficking in synaptic plasticity and neuropsychiatric disorders. Nat Rev Neurosci. 2007;8(6):413–26.View ArticlePubMedGoogle Scholar
- Lakhan SE, Caro M, Hadzimichalis N. NMDA receptor activity in neuropsychiatric disorders. Front Psychiatry. 2013;4:52.View ArticlePubMed CentralPubMedGoogle Scholar
- Monyer H, Burnashev N, Laurie DJ, Sakmann B, Seeburg PH. Developmental and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron. 1994;12(3):529–40.View ArticlePubMedGoogle Scholar
- Sans NA, Montcouquiol ME, Raymond J. Postnatal developmental changes in AMPA and NMDA receptors in the rat vestibular nuclei. Brain Res Dev Brain Res. 2000;123(1):41–52.View ArticlePubMedGoogle Scholar
- Sheng M, Cummings J, Roldan LA, Jan YN, Jan LY. Changing subunit composition of heteromeric NMDA receptors during development of rat cortex. Nature. 1994;368(6467):144–7.View ArticlePubMedGoogle Scholar
- Magnusson KR, Scruggs B, Zhao X, Hammersmark R. Age-related declines in a two-day reference memory task are associated with changes in NMDA receptor subunits in mice. BMC Neurosci. 2007;8:43.View ArticlePubMed CentralPubMedGoogle Scholar
- Zhao X, Rosenke R, Kronemann D, Brim B, Das SR, Dunah AW, et al. The effects of aging on N-methyl-D-aspartate receptor subunits in the synaptic membrane and relationships to long-term spatial memory. Neuroscience. 2009;162(4):933–45.View ArticlePubMed CentralPubMedGoogle Scholar
- Clayton DA, Browning MD. Deficits in the expression of the NR2B subunit in the hippocampus of aged Fisher 344 rats. Neurobiol Aging. 2001;22(1):165–8.View ArticlePubMedGoogle Scholar
- Magnusson KR, Brim BL, Das SR. Selective vulnerabilities of N-methyl-D-aspartate (NMDA) receptors during brain aging. Front Aging Neurosci. 2010;2:11.PubMed CentralPubMedGoogle Scholar
- Matta JA, Ashby MC, Sanz-Clemente A, Roche KW, Isaac JT. mGluR5 and NMDA receptors drive the experience- and activity-dependent NMDA receptor NR2B to NR2A subunit switch. Neuron. 2011;70(2):339–51.View ArticlePubMed CentralPubMedGoogle Scholar
- Ehlers MD. Activity level controls postsynaptic composition and signaling via the ubiquitin-proteasome system. Nat Neurosci. 2003;6(3):231–42.View ArticlePubMedGoogle Scholar
- Quinlan EM, Olstein DH, Bear MF. Bidirectional, experience-dependent regulation of N-methyl-D-aspartate receptor subunit composition in the rat visual cortex during postnatal development. Proc Natl Acad Sci U S A. 1999;96(22):12876–80.View ArticlePubMed CentralPubMedGoogle Scholar
- Philpot BD, Sekhar AK, Shouval HZ, Bear MF. Visual experience and deprivation bidirectionally modify the composition and function of NMDA receptors in visual cortex. Neuron. 2001;29(1):157–69.View ArticlePubMedGoogle Scholar
- Paoletti P, Bellone C, Zhou Q. NMDA receptor subunit diversity: impact on receptor properties, synaptic plasticity and disease. Nat Rev Neurosci. 2013;14(6):383–400.View ArticlePubMedGoogle Scholar
- Hommel JD, Sears RM, Georgescu D, Simmons DL, DiLeone RJ. Local gene knockdown in the brain using viral-mediated RNA interference. Nat Med. 2003;9(12):1539–44.View ArticlePubMedGoogle Scholar
- Lavezzari G, McCallum J, Dewey CM, Roche KW. Subunit-specific regulation of NMDA receptor endocytosis. J Neurosci. 2004;24(28):6383–91.View ArticlePubMedGoogle Scholar
- Nathanson JL, Yanagawa Y, Obata K, Callaway EM. Preferential labeling of inhibitory and excitatory cortical neurons by endogenous tropism of adeno-associated virus and lentivirus vectors. Neuroscience. 2009;161(2):441–50.View ArticlePubMed CentralPubMedGoogle Scholar
- Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelsen TS, et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science. 2014;343(6166):84–7.View ArticlePubMed CentralPubMedGoogle Scholar
- Ploski JE, Monsey MS, Nguyen T, DiLeone RJ, Schafe GE. The neuronal PAS domain protein 4 (Npas4) is required for new and reactivated fear memories. PLoS One. 2011;6(8):e23760.View ArticlePubMed CentralPubMedGoogle Scholar
- Lasek AW, Janak PH, He L, Whistler JL, Heberlein U. Downregulation of mu opioid receptor by RNA interference in the ventral tegmental area reduces ethanol consumption in mice. Genes Brain Behav. 2007;6(8):728–35.View ArticlePubMedGoogle Scholar
- Maiya R, Kharazia V, Lasek AW, Heberlein U. Lmo4 in the basolateral complex of the amygdala modulates fear learning. PLoS One. 2012;7(4):e34559.View ArticlePubMed CentralPubMedGoogle Scholar
- Mayford M, Bach ME, Huang YY, Wang L, Hawkins RD, Kandel ER. Control of memory formation through regulated expression of a CaMKII transgene. Science. 1996;274(5293):1678–83.View ArticlePubMedGoogle Scholar
- Ostedgaard LS, Rokhlina T, Karp PH, Lashmit P, Afione S, Schmidt M, et al. A shortened adeno-associated virus expression cassette for CFTR gene transfer to cystic fibrosis airway epithelia. Proc Natl Acad Sci U S A. 2005;102(8):2952–7.View ArticlePubMed CentralPubMedGoogle Scholar
- Wang D, Fischer H, Zhang L, Fan P, Ding RX, Dong J. Efficient CFTR expression from AAV vectors packaged with promoters–the second generation. Gene Ther. 1999;6(4):667–75.View ArticlePubMedGoogle Scholar
- Ghosh A, Yue Y, Duan D. Efficient transgene reconstitution with hybrid dual AAV vectors carrying the minimized bridging sequences. Hum Gene Ther. 2011;22(1):77–83.View ArticlePubMed CentralPubMedGoogle Scholar
- Choi JH, Yu NK, Baek GC, Bakes J, Seo D, Nam HJ, et al. Optimization of AAV expression cassettes to improve packaging capacity and transgene expression in neurons. Mol Brain. 2014;7:17.View ArticlePubMed CentralPubMedGoogle Scholar
- Hioki H, Kameda H, Nakamura H, Okunomiya T, Ohira K, Nakamura K, et al. Efficient gene transduction of neurons by lentivirus with enhanced neuron-specific promoters. Gene Ther. 2007;14(11):872–82.View ArticlePubMedGoogle Scholar
- Swiech L, Heidenreich M, Banerjee A, Habib N, Li Y, Trombetta J, et al. In vivo interrogation of gene function in the mammalian brain using CRISPR-Cas9. Nat Biotechnol. 2015;33(1):102–6.View ArticlePubMedGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.