Immunohistochemical detection of transgene expression in the brain using small epitope tags

Evaluation of epitope tag expression in cell extracts

HEK293T cells were transduced with lentiviral vectors encoding different tag-eGFP fusion proteins or eGFP alone (Table 1 and Fig. 1A). The amount of vector was normalized for expression based on functional titers (transducing units; TUs) as described in materials and methods. Western blot analysis confirmed a clear expression and detection of all different fusion proteins and eGFP to comparable levels. Our in-house anti-eGFP antibody [9] as well as the different commercially available antibodies detected the appropriate fusion protein with high specificity. As shown in Fig. 1, no notable proteolytic cleavage of the fusionproteins was observed. In order to avoid saturation of the western blot signal, the anti-V5 antibody had to be used at a 1:250000 dilution, indicating a very sensitive detection of V5-eGFP by this antibody. To obtain bands with comparable intensity for cell lysates overexpressing 3flag-eGFP and flag-eGFP; the former had to be diluted 500 times, showing increased detection efficiency of the 3flag tag compared to flag alone.

Evaluation of epitope tag/antibody combinations in vivo

Comparison of the transduced regions revealed that all antibodies can detect the corresponding tag-eGFP fusion protein to varying degrees, although the maximum detection levels reached by the evaluated anti-tag antibodies were in all conditions lower than the corresponding eGFP detection with our in-house anti-eGFP antibody (Fig. 2). In order to evaluate the sensitivity of the different tag/antibody combinations, we determined the ratio of tag immunopositive surface to eGFP immunopositive surface through the transduced region (Fig. 3). This quantification demonstrates that the V5/anti-V5 and HA/HA11 combinations show the highest sensitivity. The polyclonal anti-myc antibody on the other hand detects its antigen, under the conditions tested, with very low sensitivity.

Next, fluorescent immunohistochemical stainings were performed to further characterize detection sensitivity as well as to evaluate specificity at the cellular level (Fig. 4A). Panels with eGFP show primary eGFP fluorescence without additional staining, detection with the tag-specific antibodies was performed as described in materials and methods. This data confirms that all antibodies can detect the corresponding tag in vivo. Quantification of the percentage of eGFP-positive cells that stained positive using the corresponding anti-tag antibody were performed within the centre of the transduced region (Fig. 4B). In this area, where transgene expression levels are high, all but one of the evaluated tag specific antibodies detected the corresponding protein with a similar sensitivity between 70 - 90%. The polyclonal anti-myc antibody could only detect 10% of the eGFP-positive cells. Aspecific detection, i.e. presence of tag immunoreactivity that was not eGFP positive, was negligible in all cases.

Detection of firefly luciferase using epitope tagging

As proof of principle, we tested immunohistochemical detection of tagged firefly luciferase (fLuc) following locoregional overexpression in mouse striatum. Luciferase, frequently used as in vivo imaging reporter gene, can generate photons in the presence of its substrate D-luciferin and ATP. These photons can be detected by a light sensitive device. However, immunohistochemical confirmation of luciferase expression is often crucial. Although anti-luciferase antibodies are commercially available, their sensitivity for immunohistochemical detection in brain is very low [11]. As the V5 tag/anti-V5 antibody showed the highest sensitivity for V5-eGFP in vivo, we tested the detection of V5-tagged luciferase. As a control for transduced cells, V5-luciferase was co-expressed with eGFP using a T2A peptide sequence [12]. The western blot presented in Fig. 5A confirms expression of our construct. The V5-luciferase fusion protein could be detected with both the luciferase antibody and the anti-V5 antibody in cellular extracts. To evaluate the detection of V5-luciferase in vivo, we injected the lentiviral vector encoding V5-luciferase-T2A-eGFP in the striatum of mice. To assess the expression, functionality and stability of luciferase protein after lentiviral transduction, mice were imaged by bioluminescence imaging (BLI) at 4 days and 2 weeks post-injection (1.28E+06 +/- 8.30E+05 and 6.29E+05 +/- 2.65E05) (Fig. 5B). Images of both time points demonstrated a signal originating from the injected side of the brain. Animals were sacrificed 2 weeks post-injection and brains were processed for immunohistochemical analysis. In order to evaluate detection efficiency, adjacent sections were incubated with either the anti-V5 antibody, anti-luciferase antibody or the in-house anti-eGFP antibody (Fig. 5C). Despite confirmation of luciferase activity by BLI, the anti-luciferase antibody failed to detect the transgene, while the anti-V5 antibody clearly detected V5-luciferase in the transduced region visualized by eGFP.

Cloning

The different epitope tags AU-1 (6 amino acids (aa)), flag (8 aa), 3flag (22 aa), HA (9 aa), myc (10 aa) and V5 (14 aa) were N-terminally fused to eGFP (Tables 1 and 2 and Fig. 1A) by adaptor ligation in a peGFP-N1 plasmid (Clontech, Mountain View, CA, USA). All tagged eGFP constructs contained a linker sequence between tag and eGFP of RDPPVAT. To avoid eGFP expression alone, we mutated the startcodon of eGFP to ATT via a PCR based method [31]. These fusion DNA sequences, as well as eGFP alone, were subcloned into the pCHMWS plasmid backbone, an in-house lentiviral transfer plasmid (as described in [10]) in which transgene expression is under control of a CMV promoter. V5-fLuc and 3flag-fLuc, followed by a T2A eGFP sequence, were also cloned into the pCHMWS backbone. The T2A sequence allows bicistronic gene expression since the presence of this viral peptide results in equimolar expression of V5-fLuc or 3flag-fLuc and eGFP [12]. All constructs were sequence confirmed.

Lentiviral vector production

HIV-1-derived vector particles encoding the fusion proteins tag-eGFP, V5-luciferase-T2A-eGFP, 3flag-luciferase-T2A-eGFP or eGFP alone were produced, essentially as described in Ibrahimi et al. [12]. Briefly, after seeding HEK293T cells in a 10-cm dish, we performed a triple transient transfection with the respective transfer plasmids, a packaging plasmid and an envelope plasmid encoding VSV G. The production was performed in Opti-MEM I (Gibco-Invitrogen, Merelbeke, Belgium). The medium was replaced after 24 h. Cell supernatant containing lentiviral vectors was collected on day 2 and 3 post-transfection, filtered through a 0.45 μm pore size filter (Sartorius, Minisart, Göttingen, Germany) and concentrated 50-fold using a Vivaspin 15 column (Vivascience, Hannover, Germany).

Functional vector titers were determined by transducing HEK293T cells with the respective vector preparations in a 10-fold dilution series. Three days after transduction, cells were harvested and fixed in 4% paraformaldehyde (PAF) and analysed using a FACSCalibur flow cytometer (BD Biosciences, Erembodegem, Flanders) and the CellQuest software package provided with the instrument. These functional vector titers are expressed as transducing units (TUs) per ml.

Antibodies

Table 2 shows an overview of the antibodies used in the present study. These antibodies were used for detection via western blotting as well as IHC. To determine the optimal antibody dilution for immunohistochemical evaluation, three different dilutions were tested: the antibody manufacturer's recommended dilution as well as a 5-fold lower and 5-fold higher dilution. Depending on the origin of the primary antibody, goat-anti-mouse or goat-anti-rabbit secondary antibodies were used.

Cell culture and transduction

HEK293T cells were maintained in Dulbecco's modified Eagle's medium (Gibco) supplemented with 10% heat-inactivated fetal calf serum (Harlan Sera-Lab Ltd., International Medical, Brussels, Belgium) and 50 μg/ml gentamycin at 37°C in a humidified atmosphere containing 5% CO₂.

Cells were seeded in a 24 well plate at 200000 cells/well. 24 h after seeding, cells were transduced using 10⁶ TUs/ml vector. After 48 h, cells were harvested and lysed in 1% sodium dodecyl sulphate (SDS) lysis buffer containing protease inhibitor cocktail (Sigma-Aldrich, Bornem, Belgium).

Western blot analysis

Protein content of cell lysates was determined using the bicinchoninic acid (BCA) protein determination assay (Pierce Biotechnology, Rockford, IL, USA). Cell lysates were boiled in 1% SDS sample buffer containing protease inhibitor cocktail for 5 minutes (min). 2 μg of each total protein extract was resolved on a 12.5% polyacrylamide gel. Separated proteins were transferred to a polyvinylidene fluoride membrane and aspecific binding sites were blocked for 30 min in PBS supplemented with 0.1% Triton X-100 (PBST) and 5% non-fat milk. After overnight incubation at 4°C with the appropriate antibody, blots were washed 3 times with PBST. An anti-mouse beta-tubulin antibody was used as loading control. After incubation with the appropriate horseradish peroxidase-labeled secondary antibody, blots were again washed as mentioned before. Bands were visualized using enhanced chemiluminescence (Amersham Pharmacia Biotech, Little Chalfont, England).

Stereotactic injections and perfusion

Housing and handling of rats and mice were done in compliance with national guidelines; all animal procedures used were approved by the Institutional Care and Use Committee of the Katholieke Universiteit Leuven. All experiments involving viral vectors were carried out under biosafety level 2 conditions.

Eight week old female Wistar rats were used for tag-eGFP injections, while eight week old female C57BL/6-Tyrc-2J/J mice were used for BLI. The animals were anaesthetized and placed in a stereotactic head frame. After making a midline incision of the scalp, a burr hole was drilled in the appropriate location at both sites of the skull using bregma as reference. Following coordinates were used for mouse and rat respectively: anteroposterior 0.5 and 0 mm; lateral 2.0 and 2.8 mm; dorsoventral 3.0 and 5.5 mm. Three microliters (corresponding to 1.26 E+04 to 5.67 E+04 TU) of lentiviral vector supplemented with polybrene (to 4 μg/ml) was injected in rat striatum at a rate of 0.25 ml/min with a 30-gauge needle on a 10-ml Hamilton syringe. Mice were injected with 2 microliters lentiviral vector. After the injection, the needle was left in place for an additional 5 min before being slowly withdrawn from the brain (adapted from [32]). Two weeks later, animals were deeply anesthetized using an overdose of pentobarbital and transcardially perfused with saline followed by ice-cold 4% PAF in PBS. The brain was removed from the skull and post-fixed overnight in 4% PAF-PBS at 4°C.

Histology

50-mm-thick coronal brain sections were cut with a microtome (HM650V, Microm, Walldorf, Germany) and stored at 4°C in PBS with 0.1% sodium azide. The transduced area was identified based on eGFP expression using an inverted fluorescence microscope (Leica DMR optical microscope, Microsystems, Wetzlar, Germany). Sections around this injection area were used for histology with the different tag/antibody combinations.

Polyclonal in-house rabbit anti-eGFP antibody [9], a small epitope anti-tag antibody (Table 2) or anti-luciferase antibody (Promega, Madison, WI, USA) was used to detect the transgene. Chromogenic immunostaining was performed using an avidine-biotine-peroxidase complex immunostaining technique and carried out under uniform conditions. To remove endogenous peroxidase activity, sections were incubated with 3% hydrogen peroxide in PBS for 10 min. Non-specific sites were blocked by pre-treatment with 10% normal goat serum in PBST. Sections were then incubated overnight at room temperature with the appropriate primary antibody diluted in PBST with 10% goat serum as described in table 2. After being washed 3 times 5 min with PBST, the sections were incubated for 30 min at room temperature with goat anti-rabbit or anti-mouse biotinylated secondary antibody (Dako, Glostrup, Denmark) diluted at 1:300 in PBST. The sections were washed in PBST 3 times 5 min and incubated with Strept-ABC-HRP complex (Dako) for 30 min. After another wash step, detection was performed with diaminobenzidine (DAB) with H₂O₂ as substrate. The sections were coverslipped on gelatin-coated slides with DPX (Fluka, Bornem, Belgium).

For fluorescent staining, sections were incubated overnight with the appropriate primary antibodies diluted in PBST, 10% sodium azide and 10% goat serum. After three PBST rinses, sections were incubated for 2 hours at room temperature with goat anti-rabbit or goat anti-mouse IgG-Alexa 633 (diluted 1: 500, Invitrogen Molecular Probes). Sections were again rinsed with PBST and mounted on microscope slides with polyvinyl alcohol (Mowiol; Merck, La Jolla, CA, USA).

Sensitivity and specificity of the staining were quantified as described below under 'microscopy analysis'.

In vivobioluminescence imaging

BLI of fLuc activity was performed as described in Deroose et al. [11]. Shortly, mice were imaged in an IVIS 100 system (Xenogen, Alameda, CA, USA). Anaesthesia was induced in an induction chamber with 2.5% isoflurane in 100% oxygen and maintained in the IVIS imaging chamber. The mice were injected intravenously with D-luciferin (126 mg/kg, Xenogen) dissolved in PBS (15 mg/ml). Subsequently, they were placed in the prone position in the IVIS and 1 min frames were acquired until the maximum signal was reached. The data are reported as the photon flux (p/s) from a 1.6 cm² circular region of interest around the head.

Microscopy analysis

DAB staining of transgene expression in brain sections was visualized by a Leica DMR optical microscope. In order to quantify the sensitivity, we determined the ratio of transduced surface detected by the anti-tag antibody to the surface detected by the anti-eGFP antibody on the adjacent slide. Three slides through the transduced region were quantified per animal with n = 3 per condition.

Fluorescence staining was visualized by confocal microscopy using a LSM510 Laser Scanning Microscope (Zeiss, Zaventem, Belgium). To determine the proportion of transduced cells recognized by the corresponding antibody, we compared the number of cells recognized by the anti-tag antibody with the number of eGFP-positive cells. Counts were based on 12 randomly taken confocal images per animal (with n = 3) within the centre of the transduced region (as in Fig. 4A). Photos were cropped and adjusted for contrast and brightness with Adobe Photoshop version 6.0 (Adobe Systems Incorporated, San Jose, CA, USA).