Mutant NADH dehydrogenase subunit 4 gene delivery to mitochondria by targeting sequence-modified adeno-associated virus induces visual loss and optic atrophy in mice

Plasmids

To construct the fusion gene containing the mitochondrial promoter, human wild-type ND4 and epitope tag, mitochondrial DNA was extracted from human cells. Using the high fidelity of pfu Turbo DNA polymerase (Stratagene, Santa Clara, CA), the 1.4 kb mitochondrial encoded ND4 gene was cloned into the Topo TA cloning vector using a kit according to the manufacturer's directions (Invitrogen, Carlsbad, CA). A final extension step using Taq DNA polymerase was also performed. The QuikChange in vitro mutagenesis kit (Stratagene) was used to add the FLAG epitope tag with appended AGA termination codon to the 3′ end of the open reading frame of the ND4 gene to obtain ND4FLAG. Base deletions and substitutions in the reading frame ND4FLAG were corrected using the QuikChange in vitro mutagenesis kit (Stratagene). Human wild-type ND4 fused in frame with FLAG was cloned into pCDNA3 and used as a template to generate mutant human ND4 with a substitution of A for G at nucleotide 1019 with site-directed mutagenesis (Quikchange II XL site-directed mutagenesis kit; Stratagene). After confirmation with sequencing, the generated mutant ND4G1019AFLAG was then cloned into self-complementary AAV backbones under the control of the mitochondrial heavy strand promoter (HSP). The parent pTR-UF plasmid was digested with Xba I and BamHI to remove the green fluorescent protein (GFP) and neomycin resistance (neoR) genes. The Topo plasmid containing the mitochondrial ND4FLAG was also digested with Xba I and BamHI for directional cloning into the similarly digested recipient pTR-UF11 plasmid. The ND4FLAG insert was then ligated into the recipient pTR-UF11 plasmid. This gave the desired plasmid designated as pTR-UF11-mND4FLAG. Next, the hybrid CMV enhancer and chicken β-actin promoter elements (used to drive gene expression in the nucleus) was removed from pTR-UF11-ND4FLAG using KpnI and Xba I digests. We purchased complementary oligonucleotides (Invitrogen) containing the mitochondrial HSP, sequence 5′-TAA CCC CAT ACC CCG AAC CAA CCA AAC CCC AAA GAC AC-3′ with added linkers containing the KpnI restriction site at the 5′ end and Xba I at the 3′ end of HSP. The annealed double-stranded HSP oligonucleotide was directionally ligated into the pTR-UF11-ND4FLAG to give plasmid pTR-UF11-HSP-ND4FLAG. We then inserted HSP-ND4FLAG or HSP mutantND4FLAG into a self-complementary (sc) AAV backbone. The parent SC-trs-SB-smCBA-hGFP plasmid was digested with KpnI and BamHI to remove the hybrid CMV enhancer and small chicken β-actin promoter elements as well as GFP. The insert containing HSP-ND4FLAG was digested out of pTR-UF11-HSP-ND4FLAG using restriction enzymes KpnI and Xba I, and was then ligated into the recipient scAAV backbone generating the construct sc-HSP-ND4FLAG containing flanking ITRs that provide all the cis-acting sequence required for replicating, packaging, and integrating the AAV vector.

To generate an AAV directed toward the mitochondria, we added a mitochondrial targeting sequence (MTS) to the open reading frame of VP2, one of the three capsid proteins that constitute an AAV virion. This MTS-targeted VP2 was generated by linking the 23 amino acid MTS presequence of COX8, 5′ATG TCC GTC CTG ACG CCG CTG CTG CTG CGG GGC TTG ACA GGC TCG GCC CGG CGG CTC CCA GTG CCG CGC GCC AA- 3′ in frame to GFP. Eag I linkers were added to the COX8GFP insert. The parent VP2 plasmid with an Eag I site inserted at residue 138 within the VP1-VP2 overlap region was digested with Eag I. The COX8GFP was ligated into the mutant VP2 parent plasmid. This created the plasmid designated COX8GFP VP2. Construction of the mutant VP2 parent plasmid and the GFP-containing VP2 mutant plasmid without the MTS has previously been described [10].

Adeno-associated viruses

Self-complementary AAV2/sc-HSP-ND4FLAG, HSP mutantND4FLAG or a control scAAV2-GFP was produced with the plasmid cotransfection method [11]. The plasmids were packaged with the COX8GFP VP2 (mutant or wild-type human ND4) using the PXX6 helper plasmid (threefold excess) into the AAV2 recombinant virus by transfection into human 293 cells with standard procedures. Briefly, the crude iodixanol fractions were purified using the Pharmacia AKTA FPLC system, the virus was then eluted from the column with 215 mM NaCl, pH 8.0, and the rAAV peak was collected. rAAV-containing fractions were then concentrated and buffer exchanged in Alcon BSS with 0.014% Tween-20, with a Biomax 100 K concentrator (Millipore, Billerica, MA). The virus titer was then measured for DNase-resistant viral genomes with real-time PCR relative to a standard. Finally, the purity of the virus was validated with silver-stained sodium dodecyl sulfate–PAGE, assayed for sterility and lack of endotoxin, and then aliquoted and stored at −80 °C.

Cell culture and infection

Human embryonal kidney cells (293T) or mouse embryonal fibroblasts infected with AAV vectors were cultured in complete high-glucose medium (DMEM). The endogenous fluorescence of COX8GFP VP2 scAAV-infected cells was observed in live cells. The cells cultured on plastic chamber slides were reacted with MitoTracker Green according to the manufacturer's specifications (Invitrogen) and then fixed in 4% paraformaldehyde for examination with a confocal microscope.

Animals

All animal procedures were performed in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and approved by the UM IACUC. For the intraocular injection of recombinant AAV, DBA/1J mice were sedated by inhalation with 1.5% to 2% isoflurane. A local anesthetic (proparacaine HCl) was applied topically to the cornea, and then a 32-gauge needle attached to a Hamilton syringe was inserted through the pars plana. One microliter of MTS-targeted scAAV2 containing mutant or wild-type human ND4 DNA (1.01×10¹¹ VG/ml), 2.19×10¹² VG/ml) was injected in the right eye and scAAV-GFP (1.03×10¹² VG/ml) injected into the left eye. The animals were euthanized at certain time points post injection.

Electrophysiology

In brief, mice were weighed and anesthetized with intraperitoneal injections of a mixture of ketamine (80 mg/kg bodyweight) and xylazine (10 mg/kg bodyweight) and were restrained with a bite bar and a nose holder that allowed unobstructed vision. The animals were kept at a constant body temperature of 37.6 °C with a feedback-controlled heating pad. In these conditions, the eyes were wide open and in a stable position with undilated pupils pointing laterally and upward. The electroretinogram electrode had a diameter of 0.25 mm, and was made of silver wire configured to a semicircular loop with a 2-mm radius, which was placed on the corneal surface with a micromanipulator and positioned to encircle the pupil without limiting the field of view. Reference and ground electrodes were stainless-steel needles inserted under the skin of the scalp and tail, respectively. After the mice were set on the stage and before responses were recorded, a small drop of balanced saline was topically applied on the cornea to prevent drying. A visual stimulus of contrast-reversing bars (field area, 50°×58°; mean luminance, 50 cd/m²; spatial frequency, 0.05 cycles/deg; contrast, 100%; and temporal frequency, 1 Hz) was aligned with the projection of the pupil at a distance of 20 cm. The eyes were not refracted for the viewing distance, because the mouse eye has a large depth of focus due to the pinhole pupil. Retinal signals were amplified (10,000 fold) and band-pass filtered (1–30 Hz). Three consecutive responses to 600 contrast reversals each were recorded. The responses were superimposed to check for consistency and then averaged. Pattern electroretinography (PERG) is a light-adapted response. To have a corresponding index of outer retinal function, light-adapted flash electroretinography was also recorded with undilated pupils in response to strobe flashes of 20 cd/m²/s superimposed on a steady background light of 12 cd/m² and presented within a Ganzfeld bowl. Under these conditions, rod activity is largely suppressed while cone activity is minimally suppressed. To prevent bias in evaluating the peaks and troughs of PERG waveforms with reduced amplitude, the root-mean-square (RMS) voltage of the PERG signal was calculated. Response amplitude was defined as the RMS voltage over a time epoch of 0 to 350 ms after each contrast reversal according to the equation [VRMS=√ ((Σ v_i²)/n], where VRMS is the amplitude, v_i² is the voltage at each sampled time point, and n is the number of sampled points. RMS evaluation was automatically performed using a macro written in Sigma Plot language (Sigmaplot 11.2, Systat Software Inc., San Jose, CA).

In vivo imaging

In vivo high-resolution imaging of the living mouse retina was performed with spectral domain optical coherence tomography (OCT), as previously described [12]. Briefly, ketamine/xylazine anesthetized mice were restrained in a mounting tube that was fixed on a platform. Rasterscans—typically measuring 512×128 (horizontal x vertical) and 1024×64 depth scan patterns, with the fast scan in the horizontal direction—were performed for each eye. The scan length was approximately 32° for imaging mouse retinas. Acquiring high-quality OCT images took approximately 5 min for each mouse.

Polymerase chain reaction and reverse transcriptase–polymerase chain reaction

Mitochondrial and nuclear fractions were isolated from pooled ocular tissues (n=6 mice). DNA and RNA were extracted respectively from each fraction with the DNeasy Blood and Tissue Kit (Qiagen) and the RNeasy Protect Mini Kit (Qiagen). Reverse transcription was performed with the IScript One-Step RT–PCR kit (Bio-Rad). PCR was performed using primers hND4F1: 5′-ATG CTA AAA CTA ATC GTC C-3′ and hND4R1403: 5′-CTC TTG TCA TCG TCG TCC-3′ for full-length ND4FLAG of 1.4 kb; hND4F5415: 5′-TGC ATA CTC TTC AAT CAG CC-3′ and hND4R1403: 5′-CTC TTG TCA TCG TCG TCC-3′ for 500 bp ND4FLAG. The resultant PCR products were sequenced with primer hND4F5415: 5′-TGC ATA CTC TTC AAT CAG CC-3′ or hND4F1: 5′-ATG CTA AAA CTA ATC GTC C-3′.

Histology and ultrastructure

At the appropriate time points, all mice inoculated with the AAV vectors were injected with Euthasol Solution. They were then perfused by cardiac puncture with phosphate buffered solution (PBS; Na₂HPO₄ 10.9 g, NaH₂PO₄ 3.2 g, NaCl 90 g, H₂O 1000 ml; pH to 7.4) and then with fixative consisting of 4% paraformaldehyde in 0.1 M PBS buffer (pH 7.4). Whole retinas or longitudinal retinal sections were used for immunofluorescence. The following antibodies were used: Cy3-conjugated FLAG (1:100; Sigma), porin (1:100; Abcam), Thy1.2 (1:1500; Sigma), GFP (1:100; Sigma), NADH dehydrogenase subunit 4 (1:480; Abcam), and secondary antibodies that included antirat 488, antimouse Cy5, antirabbit 647, antirabbit Cy3 (1:600; Invitrogen), or antimouse immunoglobulin conjugated to immunogold (Ted Pella, Redding, CA) that was enhanced with a kit according to the manufacturer’s specifications (Ted Pella). Images were visualized with a Leica TCS (Buffalo Grove, IL) SP5 confocal microscope.

For electron microscopy, infected cultured cells or the eyes and optic nerves of infected rodents that were dissected after euthanasia and perfusion with fixative (4% paraformaldehyde, 2.5% glutaraldehyde) were further processed by immersion fixation in 2.5% glutaraldehyde, postfixed in 1% osmium tetroxide, 0.1 M sodium cacodylate-HCl buffer (pH 7.4), and 7% sucrose in the cold, and then dehydrated through an ethanol series to propylene oxide, infiltrated, and embedded in epoxy resin that was polymerized at 60 °C overnight. For immunocytochemistry, tissue specimens were postfixed in 5.0% acrolein, 0.1 M sodium cacodylate-HCl buffer (pH 7.4), and 7% sucrose and then dehydrated through an ethanol series and embedded in LR White (Ted Pella) that was polymerized at 50 °C overnight. Semithin longitudinal sections (0.5 µm) of the optic nerve head and retrobulbar nerve were stained with toluidine blue for light microscopic examination. Ultrathin sections (90 nm) were placed on nickel grids for immunocytochemistry with anti-FLAG antibodies or GFP antibodies counterstained by secondary antibodies conjugated to 5-nm immunogold. The gold particles were then amplified with a silver enhancement kit (Ted Pella). Examinations were made with a Hitachi H-7600 (Tokyo, Japan) with a tilting stage for three-dimensional rendering of the collected images or an H-7650 transmission electron microscope, operating at 200 or 80 kV, respectively.

Immunoblotting

For western blot analysis, sonicated proteins from total cellular lysates or mitochondrial pellets obtained from infected ocular tissues (n=10 mice) were electrophoresed through a 10% polyacrylamide gel and electrotransferred to a polyvinylidene fluoride membrane. The membrane was immunostained with antibodies against ND4 (1:200; Santa Cruz, Santa Cruz, CA), FLAG (1:2,000; Sigma), or NDUFS4 (1:1,000; Mitosciences, Eugene, OR) followed by either antimouse or antirabbit horseradish peroxidase (HRP)-conjugated antibodies. Immune complexes were detected using ECL reagents (HRP-conjugated antibodies) and a FUJIFilm Imaging system (Valhalla, NY).

Optic nerve areas and axon counts

One year after the viral gene injections, the optic nerves were dissected from 1 mm behind the globe to the optic chiasm. After the specimens were prepared, toluidine blue stained images were collected using an Olympus IX50 microscope, equipped with an Olympus DP7e CCD camera (Olympus America, Melville, NY). The optic nerve cross sections were manually traced for area measurements by an observer masked to the treatment. For the axon counts, 15 transmission electron micrographs were photographed at low magnification (2,500×) for each mutant ND4 injected optic nerve specimen and five micrographs for controls injected with AAV-GFP. The number of axons was then manually counted by an observer masked to the treatment agent. Axons were identified by the internalized neurotubules and axolemma surrounded by the electron-dense myelin sheath. Axons counts were expressed per mm2.

Statistical analysis

Values were expressed as means±standard deviation (SD). Data were analyzed via the Student t-test for paired data. p values of <0.05 were considered significant, and p<0.01 was highly significant.

Mitochondrial targeting sequence modified AAV does not lose infectivity

We directed the AAV virion toward the mitochondria by adding the 23 amino acid cytochrome oxidase subunit 8 (COX8) presequence fused in frame to the N-terminus of GFP into the AAV2 capsid open reading frame (Figure 1A). To direct expression of exogenous wild-type human ND4 in the organelle, we linked a mitochondrial promoter, the HSP, to the N-terminus of either wild-type human ND4 or G11778A ND4 cDNA (Figure 1B) to a scAAV backbone that contained AAV2 inverted terminal repeats (ITRs). For immunodetection, we linked a FLAG epitope to the C-terminus of G11778A ND4. The COX8 GFP coding sequence was ligated into the AAV capsid VP2 gene at a unique EAGI site at residue 138. This insertion did not substantially reduce infectivity of the recombinant AAV [13], as we achieved relatively high titers (2.2×10¹² vg/ml) for this mitochondria-targeted AAV.

Mitochondrial targeting sequence AAV-capsid localizes within mitochondria where the delivered human ND4 is translated in cultured cells

Fluorescence microscopy of live cultured cells infected with the COX8GFP VP2 MTS AAV revealed punctate and perinuclear expression of GFP (Figure 1C), suggestive of mitochondrial localization. Cells stained with a bona fide mitochondrial marker MitoTracker Green revealed a similar cellular distribution of mitochondria (Figure 1D). To determine if the MTS AAV virion entered the mitochondria, we used transmission electron microscopy where GFP labeled by silver enhanced immunogold decorated the interior of the organelle (Figure 1E). GFP immunogold was also seen within the nucleus where the AAV typically delivers its DNA. Control cells labeled only with the secondary antibody showed a background of approximately one to two gold particles per micrograph. In addition, wild-type ND4FLAG labeled by immunogold was seen only within the organelle (Figure 1F). Thus, it appears that, at least in cultured cells, a mitochondrial-encoded gene can be delivered to the interior of the organelle where the protein specified by the gene is translated.

Mutant human G11778A ND4 transmission in mice

Next, we tested for MTS AAV DNA in the mouse visual system. Here the right eyes of the mice received injections into the vitreous body of sc-HSP-ND4FLAG or sc-HSP mutantND4FLAG packaged by COX8GFP VP2 AAV. Injections of scAAV-GFP into the left eyes served as controls. To determine if the mitochondria-associated virus transferred mutant human ND4FLAG DNA, mitochondrial-enriched, cytoplasmic, and nuclear DNA fractions were extracted from the retinas and optic nerves of mice a week after intraocular viral injections. PCR for ND4FLAG gave the expected 500 kb band for the retinas and optic nerves of COX8 targeted G11778A ND4FLAG. The corresponding DNA sequences and alignment to wild-type human and mouse ND4 confirmed that the PCR products were indeed mutant human G11778A ND4, further supporting that exogenous ND4 was imported into retinal and optic nerve mitochondria by a mitochondria-targeted AAV.

Mutant human G11778A ND4 transcription in mice

To determine if the mitochondria-associated virus transfer of mutant human ND4FLAG DNA to the organelle was transcribed, mitochondrial-enriched, cytoplasmic, and nuclear RNA fractions were extracted from the retinas and optic nerves of mice a week after intraocular viral injections. RT–PCR for ND4FLAG gave the expected 500 kb band for the retinas and optic nerves of MTS-targeted G11778A ND4FLAG (Figure 2A). DNase I–treated control samples gave no RT–PCR product. The corresponding DNA sequences (Figure 2B) and alignment to wild-type human and mouse ND4 confirmed that the RT–PCR products were indeed mutant human G11778A ND4 (Figure 2C), supporting that exogenous ND4 was translated in retinal and optic nerve mitochondria following gene delivery by a mitochondria-targeted AAV.

Mutant human G11778A ND4 translation in mice

Immunoblotting of isolated optic nerve and retinal mitochondrial proteins showed that the MTS-targeted AAV directed the synthesis of mutant human ND4FLAG detected with an anti-FLAG antibody in experimental eyes but not in control eyes injected with scAAV-GFP (Figure 2D). In mutant ND4 injected eyes, ND4 (52 kDa) was overexpressed relative to the control eyes injected with GFP. The housekeeping NDUFS4 (18 kDa) protein is shown for comparison. Although ND4 cDNA and RNA were detected in the nucleus and cytoplasm, translation of ND4 can occur only within the mitochondria. This is due to the partially different genetic codes, where the TGA codon is a termination sequence for protein synthesis in the nucleus and cytoplasm (Figure 2E), but specifies the amino acid tryptophan within the mitochondria (Figure 2F).

Wild-type human ND4FLAG expression in retinal ganglion cells does not cause visual loss

Immunohistochemistry of the retinas of eyes injected with the MTS AAV containing wild-type human G11778G ND4 cDNA showed that in the RGC layer (Figure 3A) where RGCs were labeled by an antibody against Thy1.2 (Figure 3B), mutant or wild-type human ND4FLAG expression (Figure 3C) colocalized to RGCs (Figure 3D) whose axons comprised the optic nerve. Our next step was to determine that expression of wild-type human ND4 would not cause visual loss in mice. For this, we tested the visual function of eyes of animals (n=6) injected with wild-type ND4 packaged with MTS-targeted AAV. Seven weeks after AAV injections, PERGs, a sensitive measure of RGC and visual function, showed no differences in amplitude between eyes injected with wild-type human mitochondrial (m) ND4 (22.2±2.1 uV; mean±standard error) relative to uninjected eyes (21.2±1.3, n=16; Figure 3E-F).

Mutant human G11778A ND4 causes visual loss in mice

In contrast, starting at 1 month post injection, PERGs of the right eyes injected with the mutant human ND4 packaged with the MTS scAAV showed a significant decrement in RGC function (3.96±0.45 μV RMS; mean±standard error) relative to the left eyes injected with scAAV-GFP packaged with standard scAAV2 (5.16±0.45 μV RMS), p<0.04 (n=15 mice; Figure 4A). Six months post injection, this decrement between the experimental (4.43±0.45 μV RMS) and control eyes (5.85±0.46 μV RMS) was maintained, p<0.02 (Figure 4B-E). In a comparison of the 1-month PERGs and 6-month PERGs of the right eyes, we found no statistically significant difference (p=0.4). Similarly the left eyes injected with AAV-GFP had no statistically significant difference in the PERGs at 1 month relative to 6 months post injection (p=0.36). Thus, as also seen in patients with LHON, our mice with expression of G11778A mutant human ND4 had an initial drop in visual function with little propensity for spontaneous recovery.

In vivo imaging of early optic disc edema and later optic nerve head atrophy

Spectral domain optical coherence tomography (SD-OCT) of right eyes injected with the mutant human ND4 packaged with the MTS scAAV revealed swelling of the optic nerve head starting at 1 month post injection (Figure 5A) relative to the contralateral control eye injected with scAAV-GFP (Figure 5B). One year later, optic nerve head atrophy (Figure 5C) was apparent in the mutant ND4 injected eyes but not in the GFP controls (Figure 5D). At this time point, marked focal thinning of the inner retina was also apparent (Figure 5E) in an experimental eye but not observed in control eyes (Figure 5F).

Postmortem histopathology of retinal ganglion cell loss

Light microscopy of the eyes of mice euthanized a year post AAV injections confirmed the OCT findings. In the animal where OCT showed marked focal thinning of the inner retina, we found loss of the RGC and inner nuclear layers with only the outer retina remaining (Figure 6A,B). Other animals did not exhibit such severe inner retinal atrophy, but showed the more characteristic loss of cells in the ganglion cell layer (Figure 6C). In contrast, eyes injected with scAAV-GFP showed no loss of the inner retina (Figure 6D,E) or cells in the RGC layer (Figure 6F).

Postmortem histopathology of optic nerve degeneration

Postmortem analysis performed 1 year after intravitreal injections confirmed that the MTS-targeted scAAV mutant ND4 induced the hallmark optic atrophy of LHON. Relative to the normal optic nerves of contralateral control eyes that received scAAV-GFP (Figure 7A,B), the hallmark optic nerve atrophy of LHON was prominent in mouse eyes injected with MTS scAAV mutant ND4 (Figure 7C,D). Differences in the optic nerve areas between COX8 MTS delivered mutant ND4 (112,469± 4,981 μm²; mean±standard error) relative to AAV-GFP (138,300±7,208 μm²) were highly significant (p=0.004; Figure 7E). Using transmission electron micrographs to quantitate the number of optic nerve axons, we found that relative to the scAAV-GFP injected eyes with a mean of 260±33 (mean±standard deviation) axons per mm², axon counts were significantly reduced in MTS AAV mutant ND4 injected eyes with a mean of 208±24 axons per mm² (p=0.0004). A quantile plot shows that axon counts of nine mutant ND4 injected optic eyes were below the 25% quantile for the control eyes injected with AAV-GFP (Figure 7F). Thus, transmission electron microscopy confirmed that optic atrophy was due to axonal loss.

Ultrastructural analysis of scAAV-GFP injected controls revealed a full complement of normal axons enveloped by myelin (Figure 8A). In sharp contrast, axonal density was markedly reduced in the contralateral optic nerve of this same animal, 1 year after injection of scAAV mutant ND4, (Figure 8B) with empty spaces present where axons were lost. Ongoing axonal degeneration was still evident in some axons with electron dense aggregations. In a different animal, the control optic nerve injected with AAV-GFP had normal axons and myelin (Figure 8C), while the opposite optic nerve injected with scAAV mutant ND4 showed severe axonal loss with many empty spaces at foci where axons had been (Figure 8D). Even 1 year after the MTS scAAV mutant ND4 injection, some remaining axons were still in various stages of irreversible degeneration with electron dense aggregations within the axon (Figure 9A,B). Electron lucent mitochondria accumulated in swollen axons (Figure 9A). Fusion of two (Figure 9B) or three mitochondria (Figure 9C) was also evident in axons with otherwise normal appearing neurotubules.