We have produced high-titre HIV-1 green fluorescent protein-expressing lentiviral (LV) vectors pseudotyped with strain 3908 Venezuelan equine encephalitis virus glycoprotein (VEEV-G) and used them to study transduction of: (1) rat embryonic motor neuron (MN) and striatal neuron primary cultures, (2) differentiated MN cell line NSC-34 and (3) adult rat striatum. In primary neuronal cultures, transduction with VEEV-G-pseudotyped LV was more efficient and more neuronal than with vesicular stomatitis virus glycoprotein (VSV-G)-pseudotyped LV. In NSC-34 cells clear retrograde transport of VEEV-G vector particles was observed. In the striatum at the injection site, transduction with the VEEV-G vectors driven by cytomegalovirus or phosphoglycerate kinase promoters exhibited a distinct neuronal tropism with no microglial and only a minor astroglial component, superior to that obtained with VSV-G-pseudotyped LV, irrespective of the promoter used. Neuronal transduction efficiency increased over time. Distal to the injection site transduction of mitral cells in the olfactory bulb, thalamic neurons and dopaminergic neurons in the substantia nigra pars compacta was detected. This, together with observations of retrograde axonal trafficking in vitro indicates that these vectors also possess low level of retrograde neuronal transduction capability in vivo. In this study, we demonstrate both strong neurotropism as well as sustainability of expression and minimal host immune response in vivo, making the VEEV-G-pseudotyped LV vectors potentially useful for gene therapy of neurodegenerative diseases.
Tropism is the specificity of a virus for a particular host tissue/cell type, determined in part by the interaction of viral envelope glycoproteins (GP) with receptors present on the surface of the host cell.1 The restricted tropism of wild-type HIV-1 proves to be a limiting factor in therapy, but HIV particles can be produced by cells superinfected with other viruses, a process known as pseudotyping. Viruses such as Moloney murine leukaemia, herpes simplex and vesicular stomatitis virus (VSV), have been shown to confer an expanded cell tropism to lentiviral (LV) vectors due to the presence of heterologous envelope incorporated surface GPs.2, 3, 4, 5, 6, 7
The high transduction efficiency and wide cell tropism of VSV-G-pseudotyped lentivirus vectors has made this the standard against which other pseudotypes are compared. VSV-G-pseudotyped LV vectors are stable and capable of withstanding purification and concentration protocols required for their clinical use.1 Their broad cell tropism, however, do not make them amenable for targeting gene delivery to specific disease sites as they lack the ability to access difficult to reach sites (such as the central nervous system, CNS) without invasive delivery methods.
Conversely, lentiviruses pseudotyped with rabies virus GP envelopes have previously demonstrated a distinct neuronal tropism coupled with high transduction efficiency when injected in the striatum of the rodent brain.8, 9 The main characteristic feature of rabies-G is its natural retrograde transport through the CNS to the brain,8, 9, 10, 11, 12, 13 an attribute that represents the possibility of targeting gene therapy of such pseudotyped LV vectors for neurodegenerative diseases specifically to neuronal cells. In addition, it opens up the possibility of a non-invasive administration of the vector by targeting the peripheral sites of neuromuscular synapses to target MNs that can be affected in diseases such as amyotrophic lateral sclerosis and spinal muscular atrophy.
Venezuelan equine encephalitis virus (VEEV), a member of the Togaviridae, genus Alphavirus, is found in tropical and subtropical regions of South, Central and North America, where it cycles between mosquito vectors and rodent populations. The horse and other equines are also targeted by some strains, from which humans can be infected via mosquitoes. The envelope proteins of this virus have been considered an excellent candidate for the present study due to the virus natural tropism to the mammalian brain.14
Two glycoproteins E1 and E2, encoded by the 26S subgenomic RNA, form heterodimers, which in turn are grouped as 80 trimers and appear as pertrusions on the alphaviral capsid. The dual action of E1 and E2 function both in receptor attachment and membrane fusion with the ability to interact with a variety of cellular receptors allowing broad cell tropism.15
Previous studies have shown VEEV glycoprotein (VEEV-G)-pseudotyped lentiviruses,14 and various VEEV strains15, 16 to infect a wide array of cell types including lung epithelial cells, lymphocytes, neuronal cells and fibroblasts; however, these studies only demonstrated infection of undifferentiated transformed neuronal cell lines and no studies in primary embryonic CNS-derived neuronal cultures were performed.
In this study, two VEEV envelopes were used to pseudotype HIV-1-based LV vectors, both derived from the more virulent epizootic subtypes. Subtype IC strain 3908 (EpA) was isolated in 1995 during the major Venezuelan epidemic from a febrile human after which it was passaged once in C6/36 mosquito cells.17 The second subtype is termed TRD subtype IAB after being isolated from the Trinidad Donkey in 1943. These vectors were produced either with phosphoglycerate kinase (PGK) promoter or with human cytomegalovirus (CMV) promoter, both of which are known to drive higher-level constitutive expression of genes in a broader variety of mammalian cell lines.18
The pseudotyping and transduction efficiency of these LV vectors was evaluated. The most efficiently pseudotyped one was administered on embryonic primary MN and striatal neuron (SN) cultures, on NCS-34 cells in vitro (CMV promoter) and was also administered to the adult rat striatum to further assess its tropism (both PGK and CMV promoters).
Production and purification of LV vectors
Green fluorescent protein (GFP) expressing HIV-1 VEEV-G-pseudotyped vectors appeared to have a lower biological titre as opposed to VSV-G ones, which has also been previously observed by Kolokoltsov et al.14 The titre values obtained on HEK293T cells for LV vectors pseudotyped with VEEV-G strains 3908 and TRD carrying the PGK promoter were in the range of 2.8–3.7 × 108 and 2.4–3.5 × 108 TU ml−1, respectively, following 2000-fold concentration by ultracentrifugation, whereas infectivity values were found to be 3.4 × 103 and 1.4 × 103 TU ng−1 p24, respectively. Titre values for vectors carrying the CMV promoter were 1.56 × 109 TU ml−1 (unconcentrated 1.6 × 107). A HEK293T titre of 1.01 × 1010 TU ml−1 (unconcentrated 4.14 × 107) was obtained with VSV-G (Indiana Strain) LV vectors used in this study, similar to that published previously13 IC strain 3908 (EpA) is very similar to the IAB strain TRD in its gene sequence (5% divergence in the envelope proteins) and amino-acid sequence (only 12 amino-acid changes within the E1, E2, E3 and 6k genes).14 Owing to its higher infectivity value, VEEV-3908 pseudotypes were chosen for further usage, but as the differences between the two strains are minimal, this choice is arbitrary as both glycoproteins should use the same target receptor(s).
Analysis of transduction rat embryonic primary cultures
To verify the transduction efficiency in neuronal cells, the transduction levels of HIV-1 LV vectors pseudotyped with VEEV-G and VSV-G, both driven by CMV promoter, were compared in two differentiated embryonic primary neuronal cultures: MNs and SNs. These cultures are enriched in neurons (about 80 and 90% neurons at first day in vitro (DIV1), respectively, which was the time of transduction) but also contain glia, which increase with time in culture (data not shown). Each type of virus was tested with the same culture batch so as to compare differences in transduction. VSV-G pseudotypes are known to enter a broad range of cell types and were used to control for potential downstream blocks to transduction. Generally, the VEEV-G-pseudotyped HIV vectors transduced the MN cultures with an efficiency that was one-fold lower than that of the VSV-G controls, but when only the specific neuronal tropism was compared, the level of transduction of the VEEV-G pseudotype was slightly better than that of VSV-G. More specifically, at a multiplicity of infection (MOI) of 10 viral particles per cell, VEEV-G-pseudotyped vector was found to transduce 22.093±4.74% of ChAT+ve cells and 18.056±9.82% of GFAP+ve cells, whereas at an MOI of 25 it transduced 31.24±5.423% of ChAT+ve cells and 13.889±15.713% of GFAP+ve cells. VSV-G-pseudotyped vector transduced, at an MOI of 10, 22.221±0.61% of ChAT+ve cells and 52.083±20.624% of GFAP+ve cells, whereas at an MOI of 25 it transduced 26.447±4.447% of ChAT+ve cells and 50±23.57% of GFAP+ve cells (Figure 1a). The overall transduction efficiency of VSV-G-pseudotyped LV was better than for VEEV-G-pseudotyped LV at all MOIs ((factor pseudotype: F(1,8)=6.1388, P<0.05), particularly for the transduction of astrocytes (interaction between the factors pseudotype and cell type: F(1,8)=8.0139, P<0.05)).
For SN cultures (performed only at an MOI of 10) the VEEV-G-pseudotyped vector transduced 27.726±5.285%, whereas the VSV-G control vector transduced 18.323±8.342%, of DARPP32+ve cells. So for the SN cultures, where only the transduction for medium spiny neurons was assessed, the VEEV-G-pseudotyped vector showed a better neuronal tropism than the VSV-G control (Figure 1b).
These findings are consistent with the ability of VEEV to infect CNS neurons and implies that these vectors may be of use for targeting such neuronal types for gene therapy applications.
Retrograde trafficking of VEEV-pseudotyped LV particles
Pseudotyping with rabies virus GP has be shown to allow retrograde transduction of spinal cord neurons, and wild-type VEEV is also known to undergo retrograde transport. To determine whether VEEV-G psudotyping may also confer retrograde transduction to LV particles, we first determined whether these particles could undergo retrograde axonal transport, using time lapse confocal microscopy to image viral particles.19 NSC-34 cells are a mouse neuroblastoma/MN cell line that form axon-like processes when grown in the presence of low serum levels.20 VEEV-G-pseudotyped particles (CMV promoter) were produced in HEK293T cells labelled with the lipophilic dye, Vibrant DiD. This dye is then incorporated into the viral envelope at the time of budding, labelling the particles with a dye that can be visualized in the far-red spectrum. These particles were then incubated with differentiated NSC-34 cells (MOI 10, 15 min) to allow uptake of the particles, before imaging in an environmental chamber. Under these conditions clear axonal transport was observed (Figure 2a and Supplementary Movie). Although a number of particles were stationary (Figure 2a), there was clear retrograde transport of labelled particles (Figure 2a, arrow and b red track). Particle tracking analysis revealed that a large number of particles underwent retrograde transport, ranging from 30 to 130 μm in distance. Further analysis of the moving particles suggest that these particles move at a range of speeds with distributions centred around 0.2, 1.0 and 2.0 μm s−1. This is consistent with them using the same kinesin-mediated retrograde trafficking pathway previously described for tetanus toxin and canine adenovirus,19, 21 although this needs to be studied further. Somewhat surprisingly, we were also able to detect a number of particles trafficking in the anterograde direction (Supplementary Movie, Figure 2b, blue track, Figure 2c). Although less common, these events were also fast moving and able to cover long distances (Figures 2c and d).
Analysis of transduction pattern in the rat CNS
The transduction efficiency of GFP-expressing HIV-1 LV vectors pseudotyped with VEEV strain 3908 was assessed in the brains of rats by intrastriatal injections. Vectors carrying both the PGK and the CMV promoters were used. The injected volume was 4 μl of vector with a titre of 3.7 × 108 TU ml−1 transducing units for the PGK and 1.56 × 109 TU ml−1 for the CMV promoter. Collecting of brains (n=3 per time point) injected with the CMV-driven vector occurred 3 weeks post injection, whereas brains injected with PGK-driven vector were collected either 5 weeks or 8 weeks post injection to further assess stability of transduction. Immunohistochemical staining on a selection of 40 μm thick sections was performed with an anti-GFP antibody to enhance the green fluorescence signal and evaluate the extent of transduction throughout the brains of injected rats (Figures 3a–c). A distinctive pattern of transduction was detected: At 3 weeks (CMV promoter) the expression spanned ∼ 3.5 mm mediolaterally, 4 mm dorsoventrally and 4 mm anteroposteriorly around the injection site; at 5 and 8 weeks (PGK promoter) GFP expression spanned ∼2 and 2.3 mm mediolaterally and 2.3 and 3.2 mm dorsoventrally, respectively. The expression at both time points spread anteroposteriorly 3.5 mm in the striatum. Notably, GFP-expressing neurons were detected in regions distal to the striatum (Figures 3a–c) including the olfactory bulb, thalamus and substantia nigra pars compacta.
The extent of transduction at the site of injection appears to involve a smaller area of the striatum, compared with previously tested rabies virus GP-pseudotyped vectors,22 whereas the intensity of expression appears to be comparable. Dual and triple staining of this area with antibodies against GFP and NeuN and/or GFAP, revealed a distinct neuronal tropism of this pseudotype, transducing cells with a neuronal morphology/profile, whereas none or very few cells with an astrocytic morphology/profile were transduced (Figures 4a–c). The transduced cells appeared to be well localized, with only a small part of immunoreactivity from fibres rather than somas. The efficiency of transduction appears to increase between the two time points examined for the PGK promoter (Figure 4b).
GFP colocalisation with other striatal neuronal types, oligodendroglia and microglia was examined by dual staining with antibodies against GFP in conjunction with DARPP32, ChAT, Parvalbumin, RIP and ED1, respectively. Only DARPP32+ve and ChAT+ve cells were found to colocalise with GFP (shown in Supplementary Figure 1). Cells positive for GFP were observed in the olfactory bulb, in the globus pallidus, in the thalamus and in the substantia nigra. All these sites are known to send projections to and receive projections from the striatum. Within the striatum the VEEV-G-pseudotyped vector was found to transduce 20.516±4.275% of striatal cells with neuronal profile (NeuN+ve) at 3 weeks (CMV promoter), 13.21±1.49% of cells with neuronal profile (NeuN+ve) at 5 weeks and 17.73±0.75% of these cells at 8 weeks (difference between the two PGK promoter time points statistically significant: F(1,4)=10.961, P<0.05; Figure 4b). Only a minority of transduced cells presented an astroglial (GFAP+ve) profile (4.667±1.856% at 3 weeks, 6.12±0.09% at 5 weeks and 3.55±1.06% at 8 weeks; Figures 4a–c). At 3 weeks (CMV promoter) the number of tranduced astroglial cells was significantly less than the number of transduced neurons (F(1,4)=17.345, P<0.05; Figure 4a). At 8 weeks the number of transduced astroglial cells was significantly less than the number of astroglial cells transduced at 5 weeks (F(1,4)=8.773, P<0.05, Figure 4c). Striatal sections of animals injected with VSV-G-pseudotyped vector (CMV promoter) showed a transduction of 36.725±5.834% of NeuN+ve cells, and 35.337±5.745% of GFAP+ve cells (data not shown). As mentioned above 20.45±4.545% of ChAT+ve were transduced at 5 weeks, and 19.87±3.205% at 8 weeks, showing no difference because of time of incubation (Supplementary Figure 1). Retrograde transport efficiency of this pseudotyped vector was compared in more detail by single immunohistochemical staining on sections obtained from the olfactory bulb and thalamus using single GFP (Figure 3), and GFP and tyrosine hydroxylase (TH) dual staining (Supplementary Figure 1), for sections from the substantia nigra. Regardless of the promoter used, the transduced cells appeared to be diffused throughout the olfactory bulb, mainly, but not exclusively, on the mitral cell layer. A statistically significant difference in the level of transduction between the two time points considered for the PGK vector was noticed (F(1,4)=7.994, P<0.05, Figure 3d). In this area VEEV-G-pseudotyped vector transduced 1.26±0.395% of DAPI+ve cells at 3 weeks (CMV promoter), 0.792±0.256% at 5 weeks and 1.498±0.166% at 8 weeks (PGK promoter). The vector exhibited a very low level of transgene expression in the thalamus during the time points examined (at 5 weeks only a single transduced cell could be observed, whereas at 8 weeks it raised to 2–3 transduced cells per animal, where none could be observed for 3 weeks/CMV promoter; Figure 3a). Many transduced fibres in pars reticulata of substantia nigra were observed, with few TH+ve neurons from pars compacta transduced (Supplementary Figure 1c). Thalamic nuclei in which transduced cells could be observed were the ventromedial, the ventrolateral, the paracentral, the submedial, the mediodorsal, the posterior and the parafascicular nuclei. No transduced cells were observed distally from the site of injection in brains injected with VSV-G-pseudotyped vector (data not shown).
The main focus of this study was the determination of cell type tropism conferred by VEEV-G to LV vector transduction in primary CNS-derived cultures and in the rodent brain. We undertook this study in an attempt to produce a LV vector system that can efficiently transduce human and other mammalian brain cell types.
Previous work by others has established that LV vectors can be pseudotyped with the envelope proteins of other alphaviruses such as Ross River virus (RRV),23, 24, 25 Sindbis virus,26 Western equine encephalitis virus,27 Semliki Forest virus28 and most recently AURA virus.29 Although these envelopes constituted valuable components to efficiently alter the tropism of LVs, to date, only RRV LV vectors have reported high enough titres to allow transgene expression to be examined in vivo.23, 30 Here, we show that VEEV-G-pseudotyped LVs can be produced at high titres, and these vectors mediate stable and efficient in vivo transgene delivery. Furthermore, administrations of RRV-pseudotyped HIV-1 into the rat brain led to an equal level of transduction of glial cells and neurons,23 whereas stereotaxic injections of RRV-pseudotyped Feline Immunodeficiency Virus in mice mainly transduced glial cells.30 In contrast, the VEEV-G pseudotyped vector described here showed a distinct and consistent tropism for neurons both in vitro, and in vivo. Moreover, in both cases the amount of glia transduced was minor, when compared with VSV-G control. A remarkable finding is that VEEV-G-pseudotyped LV did not show any toxic effect neither to cells in vitro nor to neuropil in vivo.
To the extent of our knowledge no other envelope used for pseudotyping LV vectors shows such a strong tropism for neurons, making the VEEV-G-psudotyped LVs one the most neurotropic vectors available to date. Here, we also showed that the tropism is unaltered by the specific promoter used (PGK and CMV) and that this pseudotyped LV has the capability of being transported retrogradely. Although the retrograde transduction of VEEV-G-pseudotyped LV is considerably less efficient compared with that of previously established rabies virus GP-pseudotyped vectors,8, 22 besides an intrinsic diminished retrograde transport capability of this pseudotyped vector and its cell surface receptor laminin-binding protein,17, 31 this can be partially ascribed to the lower titre obtained, which could have led to less transduction of those parts of the striatum (the olfactory tubercle among others) that are receiving afferent connections from other structures. Despite this limitation the robust localized neuronal transduction in the striatum using this pseudotype in combination CMV promoter is superior to that seen with PGK, which is contrary to that reported previously for VSVG pseudotyped LV vectors applied to the striatum.32 Thus, this combination of pseudotype and promoter should be particularly useful in gene therapy studies for diseases such as Huntington’s, where delivery of therapeutics to striatal neurons is desirable.
In addition, VEEV-Gs, as other alphavirus GPs, possess many intriguing characteristics attractive for use in pseudotyping HIV-1 vectors. First of all, the distinctive neuronal tropism gives the VEEV-G-pseudotyped LVs the capability to deliver therapeutic transgenes in CNS and is more specific than the VSV-G-pseudotyped vectors, whose tropism is broadly mediated by ubiquitously expressed unknown receptor(s). This, together with the resistance to inactivation in the human sera and the theoretical stability in blood stream line,33 improves their versatility permitting invasive administrations as well as systemic deliveries. Second, as these vectors showed the capability of infecting both human and rodent cell lines,14 their utility might extend both for preclinical experimentation and clinical trials. When compared with VSV-G the stability of production and the lack of toxicity in producer cell lines gives them the great advantage of being easily scaled up to meet clinical needs. Moreover, the retrograde transport capability here demonstrated, albeit minimal, is still noticeable and if improved could wide the applications of VEEV-G-pseudotyped LV.
Taken together, these findings confirm VEEV-G as a good alternative to VSV-G for pseudotyping LVs and the resulting vectors as good candidates for use in a broad range of viral vector applications.
Materials and methods
Transfection-optimized human embryonic kidney cells transformed by expression of the large T antigen from SV40 virus (HEK 293T/17, ATCC, CRL-11268) were cultured in complete Dulbecco’s modified Eagles medium (DMEM, Sigma Aldrich, St Louis, MO, USA, D6546) supplemented with 10% (v/v) fetal calf serum, 1% penicillin/streptomycin (Pen/Strep, Sigma, P4333) and 1% L-glutamine (Sigma G7153). NSC-34 cells were obtained from Pam Shaw (University of Sheffield) and were cultured in DMEM supplemented with 10% fetal calf serum, 1% pen/strep, 1% L-glutamine and 1% essential ammino acids (Sigma).
Production and purification of LV vectors
Recombinant non-replicative HIV-1-based LV vectors were produced using a modified transient calcium phosphate transfection protocol based on the standard four plasmid transfection of HEK293T cells.34, 35 Briefly, 12 × 150 mm tissue culture dishes were seeded with 1.4–1.6 × 107 293T cells per dish. The next day, cells were transfected with 15 μg vector plasmid (Sin-cPPT-PGK-eGFP-WHV obtained from Professor Nicole Deglon, CEA, France or pRRLsincppt-CMV-eGFPWPRE kindly provided together with HIV packaging plasmids by Professor James Uney, University of Bristol, Bristol, UK expressing eGFP from the human PGK promoter), 15 μg of plasmid expressing the HIV-1 gag/pol gene (pMD2-LgpRRE), 3 μg of plasmid expressing HIV-1 Rev (pRSV-Rev) and 5.1 μg of the plasmid expressing the appropriate envelope GP (pMD2-VSVg, pcDNA3-VEEV.3908 or pcDNA-VEEV.TRD (obtained from Dr Robert A. Davey, University of Texas)) following the addition of 1 M CaCl2. 16 h post transfection media were replaced with fresh media supplemented with 10 mM sodium butyrate. Thirty-six hours after sodium butyrate induction, vector-containing media were collected and filtered through a 45 μm filter. The VSV-G-pseudotyped vector was used throughout as a control.
Large-scale preparations were concentrated by centrifugation at 6000 r.p.m. for 12–16 h at 4 oC (Beckman Coulter Avanti J-E, F500 rotor, Beckman Coulter, High Wycombe, UK). The pellet was then resuspended in ice-cold phosphate-buffered saline (PBS), and was further concentrated by ultracentrifugation for 90 min at 20 000 r.p.m., 10 °C (Beckman Coulter Optima L-80XP). The pellet was resuspended in 50 μl ice-cold TSSM formulation buffer (20 mM Tromethamine, 100 mM NaCl, 10 mg ml−1 sucrose and 10 mg ml−1 mannitol) every 15 min for 3 h and then centrifuged at 13 000 r.p.m. for 30 s. The supernatant was stored overnight at 4 °C along with the pellet, which was resuspended with a further 50 μl ice-cold TSSM. The resuspended pellet was centrifuged at 13 000 r.p.m. for 30 s and the two supernatants were pooled, giving a final volume of 100 μl (2000-fold concentration). Vector preparations were stored at −80 °C.
Biological activity of LV vectors carrying the GFP reporter gene was determined by flow cytometry (Becton Dickinson LSR Benchtop Flow Cytometer, Becton Dickinson, San Jose, CA, USA). Briefly, HEK 293T cells were seeded in 12-well plates at 5 × 105 cells per well. Eighteen hours after seeding the cells in a single well were quantified using a haemocytometer. Cells were transduced for 6 h with a serial dilution of the vector to be quantified in the presence of 8 μg ml−1 polybrene. Seventy-two hours post transduction the percentage of GFP-positive cells was determined by flow cytometry. The number of physical LV particles present in a preparation was estimated using a HIV-1 p24 Antigen ELISA (RETROtek from ZeptoMetrix, Franklin, MA, USA).
Preparation of rat primary neuronal cultures
Eight-well chambered slides (Lab-Tek Chambered slide system, Nalge Nunc international, Rochester, NY, USA) were coated overnight with 5 μg ml−1 poly-L-Ornithin (Sigma). After washing with distilled tissue culture water they were air-dried before coating with laminin (3 μg ml−1 in Neurobasal medium) for at least 2 h at 37 °C.
Preparation of primary MNs
All animal procedures were approved by the local Ethical Committee and performed in accordance with United Kingdom Animals Scientific Procedures Act (1986) and associated guidelines. All efforts were made to minimize the number of animals used and the suffering. The animals were housed under a 12 h light/dark cycle (light phase, 1900–0700 hours) with food and water available ad libitum. Animals were supplied by Charles River, UK. Timed pregnant female Wistar rats, on the 14th day of pregnancy (E14), were used in the experiment. The animals were killed by intraperitoneal injection of 200 mg kg−1 of sodium pentobarbitone and the embryos were immediately removed and kept in modified Hanks’ balanced salt solution (HBSS, Ca2+- and Mg2+-free, Sigma-Aldrich). Under sterile conditions each embryo was transferred into a Petri dish (Duroplan, Wertheim, Germany) containing fresh HBSS for the dissection. Six ventral portions of spinal cords were pooled in 1 ml HBSS and chopped to small pieces with a sterile scalpel. They were subsequently digested with 0.025% of trypsin (Sigma) for 10 min at 37 °C. After digestion, 100 μl DNase (1 mg ml−1 in L-15 medium, Invitrogen Gibco, Paisley, UK), 100 μl bovine serum albumin (BSA) fraction V (Sigma-Aldrich, 4% in L-15 medium) and 800 μl L-15 medium were added to the tube, mechanically dissociating the clumps with a 1 ml tip and let to settle for 3 min at room temperature. Afterwards, the supernatant was collected into a clean tube. Twenty microlitres of DNAse, 100 μl BSA and 900 μl L-15 medium were added to the remaining tissues and triturated again before settling down for another 3 min. All the supernatants were pooled and centrifuged through a BSA (4% in L-15 medium) cushion applied to the bottom of the tube at 1500 r.p.m. for 5 min.
In order to achieve a further grade of purity of MN cultures, the pellet was resuspended in 2 ml L-15 medium (phenol red-free) and divided in two separate tubes, where 1 ml layer of 10.5% Optiprep (Sigma) in L-15 medium (with phenol red) was applied to the bottom.
A second centrifugation was performed at 760 g for 15 min. The MNs containing interface were collected with 1 ml pipette and pooled into a tube containing 1 ml of MN growth medium (see Supplementary Table 1). Before seeding the cells, laminin was removed and wells were washed with PBS and half of MN growth medium was added. Cells were then seeded at a density of 100 000 cells per well (1 cm2) and cultured at 37 °C in humidified atmosphere containing 5% CO2. Half of the growth medium was changed after 24 h and subsequently every 3 days.
Preparation of primary SNs
Time-mated animals were killed at 15th day of pregnancy (E15) as described above, the dissection of the lateral ganglionic eminence was performed according to Olsson et al.36 The lateral ganglionic eminences from 12 embryos were collected in HBSS with 0.1% DNAse and 0.6% glucose. They were first trypsinised (0.25%) for 15 min at 37 °C, and then centrifuged (1200 r.p.m. for 5 min) and resuspended four times in HBSS/DNAse/glucose medium. The fragments were homogenized by repeated pipetting with 1 ml tip followed by 100 μl tip, centrifuged as before and resuspended in glutamate-free neurobasal medium containing 1% B27, 2% Pen–Strep (10 000 U ml−1, 10 000 μg ml−1), 0.5 mM L-glutamine and 15 mM KCl. Cells were plated at a density of 120 000 cells cm−2. Half of the medium was changed after 24 h and subsequently every 6 days.
LV vector transduction of rat primary cells
Rat primary cell cultures of MNs (1 × 105 cells per well) and SNs (1.2 × 105 cells per well) were plated in eight-well chambered slides. Transduction of rat primary neurons was carried at DIV1 with VSV-G and VEEV-G pseudotyped concentrated LV vector stocks. MN cultures were infected at MOI of 10 and 25. Striatal neurons were infected with each LV vector stock at MOI 10. Transduction experiments were carried out in duplicate for each MOI for both LV vectors. Both types of primary neuronal culture were incubated with vector stocks in 300 μl conditioned culture medium for 6 h at 37 °C in humidified atmosphere containing 5% CO2. After 6 h, medium was replaced with 300 μl of fresh conditioned cultured medium and cells were incubated for a further 3 days at 37 °C with 5% CO2. The cells were then fixed and immunostained as described below.
Cells were fixed with 2% paraformaldehyde (Sigma) diluted in PEM/microtubules stabilization buffer (0.2 M PIPES, 10 mM MgCl2, 10 mM EGTA, pH 7.2) for 20 min in the dark, washed with PBS and permeabilised with PBS containing 2% BSA and 0.1% of Triton-X 100 (Sigma) for 8–10 min. Blocking of nonspecific binding sites was achieved by 1 h incubation in PBS with 2% BSA. Antibodies to NeuN (1:200, MAB377, Millipore, Watford, UK), DARPP32 (1:1000, ab40801, Abcam, Cambridge, UK), GFAP (1:800, MAB360, Millipore) and ChAT (1:100 AB144P, Millipore) were diluted in the same buffer and placed in the wells overnight at 4 °C. After a short blocking, cells were incubated 3 h at room temperature with secondary either donkey or goat antibodies anti-mouse and anti-rabbit, conjugated with alexafluor 594 or 647 (1:1000, Invitrogen). Following washes with PBS the wells’ walls were removed from the slides and the coverslips applied using ProLong anti-fade aqueous mounting medium (Invitrogen, Paisley, UK).
Imaging of LV trafficking
Fluorescently labelled LV particles were produced as described above, with the addition of a labelling step. Eighteen hours post transfection, cells were incubated with 3.75 μM Vibrant DiD cell labelling solution (Molecular Probes, Paisley, UK) in Opti-MEM (Invitrogen). After 2 h, media was replaced with DMEM containing 2% FBS and 10 mM sodium butyrate. Virus was then harvested as described above. NSC-34 cells were plated onto imaging dishes (Mattek, Ashland, MA, USA) coated with 100 g ml−1 poly-D-lysine, 10 g ml−1 laminin and fibronectin (10 000 cells per dish), in DMEM containing 10% FBS and 1% FBS essential amino acids (Gibco, Paisley, UK). Twenty four hours after plating media was replaced with DMEM containing 1% FBS and 1% essential amino acids to induce differentiation, and experiments were performed 72–96 h post differentiation. Live imaging was performed as described in Salinas et al.19 Briefly, cells were incubated with labelled LV particles (MOI 10) for 15 min, before cells were washed once and imaging media added. Dishes were then imaged using a TCS SP5 II confocal microscope with environmental chamber. Images were taken every 5 s using the 633 nm laser, with the photo multiplier tube detection range of 650–800 nm, as DiD fluoresces in the far-red spectrum. Images were processed using ImageJ (Rasband WS, ImageJ, U.S. National Institutes of Health, USA. http://imagej.nih.gov/ij/) and images enhanced using a 2.0 pixel median filter and brightness/contrast, and particle analysis performed using MTrackJ and multiple Kymograph plugins. Images were then rendered and pseudocoloured using GIMP (GNU Image Manipulation Programme; http://www.gimp.org).
Analysis of transduction pattern in the rat CNS
A total of six male Wistar rats, weighing 200–250 g at the start of the experiment were used. Before surgery, all animals were deeply anesthetized by inhalation of a mixture of 3 l oxygen and 3.5% isoflurane (Merial, Parramata, NSW, Australia) and then received systemic analgesia. They were placed in a stereotactic frame (Taxic-6000, World Precision Instruments, Hitchin, UK) with the nose bar set at +3.3 mm. The anaesthetic mixture was changed to 1 l oxygen and 2.2% isoflurane for the remainder of the operation. The scalp was cut and retracted to expose the skull. Craniotomy was performed by drilling directly above the target region, to expose the pial surface. One single injection was directed into the right striatum using the stereotactic coordinates relative to bregma: anteroposterior, 0.5 mm; mediolateral, 3.0 mm; dorsoventral, 5.0 mm. Animals received 4.0 μl of pseudotyped LV vectors with a biological titre of 1.56 × 109 TU ml−1 (CMV) and 3.7 × 108 TU ml−1 (PGK), by a 32G blunt needle using an infusion pump (UltramicropumpIII and Micro4 Controller, World Precision Instruments) at a flow rate of 0.2 μl min−1 over 20 min. Needle was then allowed to remain on site for an additional 5 min before slow retraction. At the completion of surgery, the skin was sutured.
Brains were analysed at three different time points: 3 weeks (CMV), 5 weeks and 8 weeks (PGK) post injection. Rats were euthanized by intraperitoneal injection of an 200 mg kg−1 of sodium pentobarbitone, transcardially perfused with 50 ml saline (0.9% w/v NaCl) plus heparin, followed by 250 ml of 4% paraformaldehyde (Sigma) in PBS. Brains were removed and post fixed for 4 h in 4% paraformaldehyde, followed by cryoprotection in 10% glycerol and 20% sucrose in PBS for at least 72 h. Brains were subsequently embedded and frozen in OCT (Surgipath FSC22, Leica Microsystems, Wetzlar, Germany). Forty micron coronal sections from the olfactory bulb, through the entire striatum, globus pallidus and substantia nigra were cut using a cryostat (Leica Microsystems), mounted on poly-L-lysine coated slides and stored at −20 °C.
Immunohistochemistry of brain sections
Immunohistochemistry was performed on slide mounted brain sections. Tissue was permeabilised and blocked for 1 h with PBS containing 10% donkey serum and 0.1% Triton-X 100. Antibodies to GFP (1:500, ab290, Abcam), NeuN (1:100, MAB377, Millipore), DARPP32 (1:1000, ab40801, Abcam), GFAP (1:800, MAB360, Millipore), RIP (1:1000, MAB1580, Millipore), ED1 (1:100, ab31630, Abcam) and TH (1:2000, MAB318, Millipore) were diluted in the same buffer and placed on sections for 48 h at 4 °C. Sections were then blocked for 30 min before incubating for 3 h at room temperature with donkey anti-mouse, anti-goat or anti-rabbit secondary antibodies conjugated with alexafluor 488, 594 and/or 647 (1:500). Sections were coverslipped with ProLong antifade aqueous mounting medium (Invitrogen) in presence of DAPI (0.05 μg ml−1, Sigma) to label nuclear DNA.
Double and triple fluorescent-labelled cells were acquired under a confocal laser-scanning microscope (TCS SP5 II, Leica Microsystems). For each well 10 randomly chosen square fields with sides measuring 125 μm were counted with a computer-assisted imaging programme (ImageJ). Specific MN transduction efficiency in embryonic primary MN cultures was assessed as the percentage of double-positive GFP/ChAT cells on the total of ChAT+ve cells. Specific astrocytic transduction efficiency was assessed as percentage of double-positive GFP/GFAP on the total of GFAP+ve cells. Selected medium spiny neurons transduced were quantified as the percentage of double-positive GFP/DARPP32 on the total of DARPP32+ve cells.
Images from single fluorescent-labelled brain sections (olfactory bulb and thalamus) were taken under an epifluorescent microscope (Eclipse 80i, Nikon, Tokyo, Japan) equipped with a motorized stage for automated tiling of adjacent fields (Prior Scientific, Cambridge, UK). For olfactory bulb images were taken using × 10 objective (NA 0.45), whereas thalamic images were taken using a × 4 objective (NA 0.2). MN cultures, SN cultures, double and triple fluorescent-labelled sections (striatum and substantia nigra) were acquired under a confocal laser-scanning microscope (TCS SP5 II). For cell counts, a series of sections through the olfactory bulb, the forebrain and midbrain were used for immunostaining. The number of immunostained cells in each brain region was counted with a computer-assisted imaging programme (ImageJ). For characterization of striatal cells around vector injection sites, representative sections were used for double and triple immunofluorescence histochemistry. In individual animals, the number of immunostained cells in the fields of interest ( × 40, NA 0.85 objective) were counted with the imaging programme. Four to six sections obtained from each brain region of individual animals were used for cell counts, and the average per animal was calculated.
Analysis of variance was used for statistical comparisons. Post-hoc analysis was Tukey Honestly Significant Difference. Analyses were carried out using R: A Language and Environment for Statistical Computing (R Development Core Team, R Foundation for Statistical Computing, Austria. http://www.R-project.org). Data are expressed as mean±s.d. values for embryonic primary cultures results and as mean±s.e.m. (n=3) for in vivo results.
Bartz SR, Rogel ME, Emerman M . Human immunodeficiency virus type 1 cell cycle control: Vpr is cytostatic and mediates G2 accumulation by a mechanism which differs from DNA damage checkpoint control. J Virol 1996; 70: 2324–2331.
Canivet M, Hoffman AD, Hardy D, Sernatinger J, Levy JA . Replication of HIV-1 in a wide variety of animal cells following phenotypic mixing with murine retroviruses. Virology 1990; 178: 543–551.
Chesebro B, Wehrly K, Maury W . Differential expression in human and mouse cells of human immunodeficiency virus pseudotyped by murine retroviruses. J Virol 1990; 64: 4553–4557.
Cronin J, Zhang XY, Reiser J . Altering the tropism of lentiviral vectors through pseudotyping. Curr Gene Ther 2005; 5: 387–398.
Lusso P, di Marzo Veronese F, Ensoli B, Franchini G, Jemma C, DeRocco SE et al. Expanded HIV-1 cellular tropism by phenotypic mixing with murine endogenous retroviruses. Science 1990; 247: 848–852.
Spector DH, Wade E, Wright DA, Koval V, Clark C, Jaquish D et al. Human immunodeficiency virus pseudotypes with expanded cellular and species tropism. J Virol 1990; 64: 2298–2308.
Zhu ZH, Chen SS, Huang AS . Phenotypic mixing between human immunodeficiency virus and vesicular stomatitis virus or herpes simplex virus. J Acquir Immune Defic Syndr 1990; 3: 215–219.
Mazarakis ND, Azzouz M, Rohll JB, Ellard FM, Wilkes FJ, Olsen AL et al. Rabies virus glycoprotein pseudotyping of lentiviral vectors enables retrograde axonal transport and access to the nervous system after peripheral delivery. Hum Mol Genet 2001; 10: 2109–2121.
Wong LF, Azzouz M, Walmsley LE, Askham Z, Wilkes FJ, Mitrophanous KA et al. Transduction patterns of pseudotyped lentiviral vectors in the nervous system. Mol Ther 2004; 9: 101–111.
Federici T, Kutner R, Zhang XY, Kuroda H, Tordo N, Boulis NM et al. Comparative analysis of HIV-1-based lentiviral vectors bearing lyssavirus glycoproteins for neuronal gene transfer. Genet Vaccines Ther 2009; 7: 1.
Kato S, Inoue K, Kobayashi K, Yasoshima Y, Miyachi S, Inoue S et al. Efficient gene transfer via retrograde transport in rodent and primate brains using a human immunodeficiency virus type 1-based vector pseudotyped with rabies virus glycoprotein. Hum Gene Ther 2007; 18: 1141–1151.
Mentis GZ, Gravell M, Hamilton R, Shneider NA, O’Donovan MJ, Schubert M . Transduction of motor neurons and muscle fibers by intramuscular injection of HIV-1-based vectors pseudotyped with select rabies virus glycoproteins. J Neurosci Methods 2006; 157: 208–217.
Mitrophanous K, Yoon S, Rohll J, Patil D, Wilkes F, Kim V et al. Stable gene transfer to the nervous system using a non-primate lentiviral vector. Gene Ther 1999; 6: 1808–1818.
Kolokoltsov AA, Weaver SC, Davey RA . Efficient functional pseudotyping of oncoretroviral and lentiviral vectors by Venezuelan equine encephalitis virus envelope proteins. J Virol 2005; 79: 756–763.
Greene IP, Paessler S, Austgen L, Anishchenko M, Brault AC et al. Envelope glycoprotein mutations mediate equine amplification and virulence of epizootic venezuelan equine encephalitis virus. J Virol 2005; 79: 9128–9133.
Phillpotts RJ, Jones LD, Howard SC . Monoclonal antibody protects mice against infection and disease when given either before or up to 24 h after airborne challenge with virulent Venezuelan equine encephalitis virus. Vaccine 2002; 20: 1497–1504.
Ludwig GV, Kondig JP, Smith JF . A putative receptor for Venezuelan equine encephalitis virus from mosquito cells. J Virol 1996; 70: 5592–5599.
Zufferey R, Donello JE, Trono D, Hope TJ . Woodchuck hepatitis virus posttranscriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J Virol Apr 1999; 73: 2886–2892.
Salinas S, Bilsland LG, Henaff D, Weston AE, Keriel A, Schiavo G et al. CAR-associated vesicular transport of an adenovirus in motor neuron axons. PLoS Pathog 2009; 5: e1000442.
Cashman NR, Durham HD, Blusztajn JK, Oda K, Tabira T, Shaw IT et al. Neuroblastoma x spinal cord (NSC) hybrid cell lines resemble developing motor neurons. Dev Dyn 1992; 194: 209–221.
Deinhardt K, Berninghausen O, Willison HJ, Hopkins CR, Schiavo G . Tetanus toxin is internalized by a sequential clathrin-dependent mechanism initiated within lipid microdomains and independent of epsin1. J Cell Biol 2006; 174: 459–471.
Carpentier DC, Vevis K, Trabalza A, Georgiadis C, Ellison SM, Asfahani RI et al. Enhanced pseudotyping efficiency of HIV-1 lentiviral vectors by a rabies/vesicular stomatitis virus chimeric envelope glycoprotein. Gene Ther 2012; 19: 761–774.
Jakobsson J, Nielsen TT, Staflin K, Georgievska B, Lundberg C . Efficient transduction of neurons using Ross River glycoprotein-pseudotyped lentiviral vectors. Gene Ther 2006; 13: 966–973.
Kahl CA, Marsh J, Fyffe J, Sanders DA, Cornetta K . Human immunodeficiency virus type 1-derived lentivirus vectors pseudotyped with envelope glycoproteins derived from Ross River virus and Semliki Forest virus. J Virol 2004; 78: 1421–1430.
Sharkey CM, North CL, Kuhn RJ, Sanders DA . Ross River virus glycoprotein pseudotyped retroviruses and stable cell lines for their production. J Virol 2001; 75: 2653–2659.
Morizono K, Bristol G, Xie YM, Kung SK, Chen IS . Antibody-directed targeting of retroviral vectors via cell surface antigens. J Virol 2001; 75: 8016–8020.
Poluri A, Ainsworth R, Weaver SC, Sutton RE . Functional pseudotyping of human immunodeficiency virus type 1 vectors by Western equine encephalitis virus envelope glycoprotein. J Virol 2008; 82: 12580–12584.
Strang BL, Takeuchi Y, Relander T, Richter J, Bailey R et al. Human immunodeficiency virus type 1 vectors with alphavirus envelope glycoproteins produced from stable packaging cells. J Virol 2005; 79: 1765–1771.
Froelich S, Tai A, Kennedy K, Zubair A, Wang P . Pseudotyping lentiviral vectors with AURA virus envelope glycoproteins for DC-SIGN-mediated transduction of dendritic cells. Hum Gene Ther 2011; 22: 1281–1291.
Kang Y, Stein CS, Heth JA, Sinn PL, Penisten AK, Staber PD et al. In vivo gene transfer using a nonprimate lentiviral vector pseudotyped with Ross River Virus glycoproteins. J Virol 2002; 76: 9378–9388.
Malygin AA, Bondarenko EI, Ivanisenko VA, Protopopova EV, Karpova GG, Loktev VB . C-terminal fragment of human laminin-binding protein contains a receptor domain for venezuelan equine encephalitis and tick-borne encephalitis viruses. Biochemistry 2009; 74: 1328–1336.
de Almeida LP, Zala D, Aebischer P, Déglon N . Neuroprotective effect of a CNTF-expressing lentiviral vector in the quinolinic acid rat model of Huntington’s disease. Neurobiol Dis 2001; 8: 433–446.
Strauss JH, Strauss EG . The alphaviruses: gene expression, replication, and evolution. Microbiol Rev 1994; 58: 491–562.
Mochizuki H, Schwartz JP, Tanaka K, Brady RO, Reiser J . High-titer human immunodeficiency virus type 1-based vector systems for gene delivery into nondividing cells. J Virol 1998; 72: 8873–8883.
Kutner RH, Zhang XY, Reiser J . Production, concentration and titration of pseudotyped HIV-1-based lentiviral vectors. Nat Protoc 2009; 4: 495–505.
Olsson M, Campbell K, Wictorin K, Björklund A . Projection neurons in fetal striatal transplants are predominantly derived from the lateral ganglionic eminence. Neuroscience 1995; 69: 1169–1182.
We would like to thank our ex-MSc students Petros Patsali, Daniel M Lipiski and Joanne Crowe for technical contributions to this study. Dr Robert A. Davey, University of Texas for giving us the VEEV-G TRD and 3908 plasmids. Professor Richard Reynolds and Dr Owain Howell for the use of and help with fluorescence microscopy to document histological staining of brain sections. We would also like to thank Dr Egle Solito and Enrico Cristante for letting us use several antibodies. This work was partly supported by Imperial College London starting funds to NDM and a Seventh Framework Programme European Research Council Advanced Grant, no: 23314 to NDM supporting AT, IE, JH and SME.
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on Gene Therapy website
About this article
Cite this article
Trabalza, A., Georgiadis, C., Eleftheriadou, I. et al. Venezuelan equine encephalitis virus glycoprotein pseudotyping confers neurotropism to lentiviral vectors. Gene Ther 20, 723–732 (2013). https://doi.org/10.1038/gt.2012.85
- viral vector
Tropism, intracerebral distribution, and transduction efficiency of HIV- and SIV-based lentiviral vectors after injection into the mouse brain: a qualitative and quantitative in vivo study
Histochemistry and Cell Biology (2017)
Specific Retrograde Transduction of Spinal Motor Neurons Using Lentiviral Vectors Targeted to Presynaptic NMJ Receptors
Molecular Therapy (2014)