Abstract
Suprachoroidal nonviral gene therapy with biodegradable poly(β-amino ester) nanoparticles (NPs) provides widespread expression in photoreceptors and retinal pigmented epithelial (RPE) cells and therapeutic benefits in rodents. Here, we show in a human-sized minipig eye that suprachoroidal injection of 50 μl of NPs containing 19.2 μg of GFP expression plasmid caused GFP expression in photoreceptors and RPE throughout the entire eye with no toxicity. Two weeks after injection of 50, 100, or 200 μl, there was considerable within-eye and between-eye variability in expression that was reduced 3 months after injection of 200 μl and markedly reduced after three suprachoroidal injections at different locations around the eye. Reduction of bacterial CpG sequences in the expression plasmid resulted in a trend toward higher expression. These data indicate that nonviral suprachoroidal gene therapy with optimized polymer, expression plasmid, and injection approach has potential for treating photoreceptors throughout the entire retina of a human-sized eye.
Widespread photoreceptor transfection in human-sized eyes shows the feasibility of suprachoroidal nonviral gene therapy.
INTRODUCTION
There has been considerable progress in ocular gene therapy using adeno-associated viral (AAV) vectors. Subretinal injection of an AAV2 vector expressing Rpe65 improved mobility in some patients with retinal degeneration due to biallelic mutations in Rpe65 (1), which has led to approval of voretigene Neparvovec-rzyl by the US Food and Drug Administration. However, since that approval, it has been observed that some patients treated with subretinal injection of voretigene Neparvovec-rzyl have developed perifoveal chorioretinal atrophy (2). The mechanism of this toxicity is not known, but the delayed onset and nature of the changes suggest the possibility of a late immune response. While it is necessary to continue to study the long-term benefits and risks of viral ocular gene transfer and find ways to minimize the latter, it is prudent to develop nonviral ocular gene transfer approaches.
Poly(β-amino ester)s (PBAEs) with primary, secondary, and tertiary amines and ester bonds can compact expression plasmid DNA into nanoparticles (NPs) that enter cells by endocytosis (3) and are hydrolytically degradable. A substantial number of the NPs escape from endosomes and degrade within minutes to hours, allowing expression plasmids to enter the nucleus and begin transgene expression (4–6). Relatively fast degradability of new biocompatible gene delivery materials is especially important for retinal gene therapy as recent studies have highlighted potential toxicity concerns of conventional materials, such as lipid NPs, for application in the retina (7, 8). Suprachoroidal injections of PBAE NPs containing a green fluorescent protein (GFP) expression plasmid result in widespread expression of GFP in photoreceptors and retinal pigmented epithelial (RPE) cells of rats (9); however, it is important to know the extent and level of GFP expression that occurs after suprachoroidal injection of PBAE NPs in eyes that are closer in size to human eyes and that may have more substantial transport limitations. We selected minipig eyes for these experiments because, in addition to being closer in size to human eyes, the minipig retina is similar to the human retina in that it is holangiotic and has an area centralis with a high density of cone photoreceptors analogous to the primate macula (10).
RESULTS
NP characterization
Following NP fabrication, lyophilization, storage at −20°C, and resuspension in sterile water, NP properties were evaluated. Assessment of NP size and surface charge revealed a z-average hydrodynamic diameter of 443 nm and a zeta potential of +29.4 mV (Fig. 1A). Transmission electron microscopy (TEM) showed that morphology was spherical for the electrostatically formed polyplex NPs and that dried particle size was similar to hydrated particle size (Fig. 1B). Gel electrophoresis studies demonstrated that the NPs had complete binding and encapsulation of DNA in the NPs (Fig. 1C). The size of these electrostatically formed NPs and the complete encapsulation of DNA plasmids allows high DNA loading per particle, calculated as 250 ± 20 kilo–base pair (kbp) DNA encapsulated in each particle on average, or 80 ± 5 plasmids of 3151 bp (fig. S1), which may be advantageous for in vivo gene therapy. The discrepancy in the particle size as measured by dynamic light scattering (DLS), TEM, and NP tracking analysis (NTA; fig. S1A) is due to the polydispersity of the particle population; while most of the particles, counted as a number average as by TEM and NTA, are relatively small (<200 nm), the less common larger particles bias the z-average calculated via intensity-averaged size by DLS.
Fig. 1. PBAE NP characterization.
PBAE NPs were prepared by being lyophilized, stored at −20°C, and then resuspended in sterile water as they were for in vivo studies. (A) NPs were assessed for hydrodynamic diameter (particle size) and zeta potential (ZP) (surface charge) in aqueous conditions via DLS. Bars represent the mean ± SE of the z-average diameter and the zeta potential from three independently prepared batches. (B) NP size and shape were visualized via TEM. (C) Gel electrophoresis showed complete binding of DNA in the NPs.
Retinal appearance after suprachoroidal injection of PBAE NPs containing a GFP expression plasmid
Experiments were done in Göttingen and Yucatan minipigs. Fundus photographs were taken before and after suprachoroidal injection of PBAE NPs. Representative before and after fundus photographs of Göttingen and Yucatan minipigs are shown in fig. S2 (A to D). Indirect ophthalmoscopy was performed at baseline before and after injection and before euthanasia to observe the entire extent of the retina. No visible changes in the retina were observed on fundus photographs or by indirect ophthalmoscopy at 2 or 12 weeks after suprachoroidal injections.
Distribution of GFP expression after suprachoroidal injection of PBAE NPs containing pCAG-GFP-Z1 expression plasmid in minipig eyes
In previously reported experiments (9), suprachoroidal injections were done with a commercially available 4733-bp GFP expression plasmid, pEGFP-N1, in which the cytomegalovirus (CMV) promoter drives expression of GFP. The 4021-bp pCAG-GFP-Z1 was generated by replacing the CMV promoter with the CAG promoter and excising extraneous bacterial sequences but maintaining those allowing propagation and selection in bacteria. Two weeks after suprachoroidal injection of 50 μl containing 19.2 μg of pCAG-GFP-Z1 DNA in PBAE NPs 4.5 mm posterior to the limbus, minipig eyes were fixed, run through a sucrose gradient, and frozen in embedding medium. Starting at the anterior-most part of the eyecup, 35-μm transverse sections were cut proceeding as far posterior as possible. The section through the equator, which has the largest diameter, was identified, and then the distance anterior or posterior to the equator was determined for all other sections (Fig. 2A). Figure 2B shows fluorescence microscopy of sections 2.87 mm anterior to the equator, at the equator, and 8.12 mm posterior to the equator. The fluorescence extends around the entire circumference of the eye at each of the three locations. Higher magnification of the equatorial section immunohistochemically stained with anti-GFP antibody shows that the fluorescence is due to GFP expression predominantly in photoreceptor inner and outer segments (Fig. 2C). This indicates that 2 weeks after a single anterior suprachoroidal injection of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-Z1, GFP expression occurs in photoreceptors around the entire circumference of the eye and extends as far posterior as can be determined by this technique. Staining of these sections with hematoxylin and eosin showed normal-appearing retina with no inflammatory cells (fig. S2, E and F).
Fig. 2. Suprachoroidal injection of PBAE NPs containing a GFP expression plasmid in minipigs causes widespread expression of GFP in photoreceptors.
(A) Schematic showing the manner in which transverse sections were used to localize expression relative to the equator of the eye. (B) Two weeks after suprachoroidal injection of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-Z1 in a Göttingen minipig, transverse ocular sections 2.87 mm anterior to the equator, at the equator, or 8.12 mm posterior to the equator each showed GFP fluorescence around the entire circumference of the eye. (C) At higher magnification, immunofluorescent staining for GFP demonstrated that the fluorescence is due to expression of GFP in photoreceptor inner and outer segments.
Effect of increasing the volume/dose and/or time after injection of NPs
The eye is a closed compartment, and any injection of fluid into the eye results in a transient increase in intraocular pressure, which normalizes as aqueous humor exits through the outflow channels of the eye. In humans, the most common volume of medication injected into the vitreous cavity or suprachoroidal space is 50 μl, but injections of 100 μl are tolerated in most patients unless outflow facility is limited by glaucoma. If the rate of injection is very slow, even larger volumes can be injected without increasing intraocular pressure above ocular perfusion pressure. Similar to the previous experiment, 2 weeks after suprachoroidal injection of 50 μl of PBAE NP containing 19.2 μg of pCAG-GFP-Z1, GFP fluorescence was seen around the entire circumference of the eye in all transverse sections up to about 9 mm posterior to the equator, which for technical reasons is about the limit for getting an intact circular section (Fig. 3, A and B). The distribution of GFP expression was similarly widespread 2 weeks after injection of 100 μl of NPs containing 38.4 μg of pCAG-GFP-Z1 (Fig. 3, C and D) or 200 μl of NPs containing 76.8 μg of pCAG-GFP-Z1 (Fig. 3, E and F). Since fluorescence microscopy is very sensitive but not quantitative, these experiments could not determine if there were differences in level of GFP expression after injection of the different volumes of NPs.
Fig. 3. Comparison of GFP expression 2 weeks after suprachoroidal injection of 50, 100, or 200 μl of NP containing pCAG-GFP-Z1.
(A to F) Yucatan minipigs were given a suprachoroidal injection of 50, 100, or 200 μl of PBAE NP containing pCAG-GFP-Z1 (0.38 μg/μl). Two weeks after injection of each of the volumes, GFP fluorescence was seen around the entire circumference of the eye sections at the equator and those posterior to the equator.
To quantify and better localize expression, GFP protein levels were measured by enzyme-linked immunosorbent assay (ELISA) in retina and RPE/choroid at several locations throughout the eye. A 7-mm trephine was used to collect circular punches of retina and RPE/choroid at seven locations: (i) posterior nasal, (ii) posterior temporal, (iii) temporal, (iv) superior nasal, (v) superior temporal, (vi) inferior nasal, and (vii) inferior temporal (fig. S3, A and B). Before trephining, the translucent retina was seen overlying the darkly pigmented RPE (fig. S3C). After removal of retinal samples, the darkly pigmented RPE was seen more clearly in the circles where the retina had been removed (fig. S3D). After removal of RPE/choroid samples, sclera was seen in the circles (fig. S3, E and F). ELISA was used to measure GFP protein/mg total protein in each of the samples.
Figure 4 (A to C) shows GFP protein/mg total protein in each of the seven retinal locations in three eyes at 2 weeks after suprachoroidal injection of 50 μl of PBAE NP containing 19.2 μg of pCAG-GFP-Z1. The injections were done superiorly in the 11:00 or 12:00 meridians at an anterior location, 4.5 mm posterior to the limbus, and GFP expression was measurable at all locations of the retina except at one location in one of the eyes. The level of expression was not consistently greater in regions of the retina in closest proximity to the injection site, and in one eye, it was substantially higher in the region of retina furthest from the injection site (Fig. 4B). There was no consistent pattern regarding differences in GFP expression at different locations within an eye or between different eyes, but detectable expression occurred in all parts of the retina after a single suprachoroidal injection. Graphical display of each of the data points for each eye provides a better indication of the within-eye and between-eye variability in regional GFP levels (Fig. 4D). In two of the eyes, Eye 1-2w and Eye 3-2w, the means (13.6 and 13.7 pg/mg protein) and range of the seven measurements were similar, despite lack of correspondence between the eyes in values from the same location. Compared with these two eyes, a third eye, Eye 2-2w, showed a mean GFP protein level that was more than ninefold higher (126.1 pg/mg protein) and higher range of values.
Fig. 4. GFP protein levels in different regions of retina after a single suprachoroidal injection of 50 μl of NP containing 19.2 μg of pCAG-GFP-Z1.
Three Yucatan minipigs were given a suprachoroidal injection of 50 μl of PBAE NP containing pCAG-GFP-Z1. Two weeks after injection, GFP protein was measured by ELISA in punches of retina. (A to C) The level of GFP in pg/mg total protein is shown at sample locations in each of the three eyes injected. (D) A dot plot of all values in each eye shows the mean and illustrates in-eye and between-eye variability. ND, not detectable; NC, not collected.
Compared with GFP protein levels 2 weeks after suprachoroidal injection of 50 μl of PBAE NPs (19.2 μg of pCAG-GFP-Z1), those seen 12 weeks after injection were comparable (Fig. 5A). In general, there was correlation between GFP levels in retina and those in RPE/choroid because eyes with higher ranges of expression in retina tended to have higher ranges of expression in RPE/choroid (Fig. 5, A and B). Figure 5C shows GFP protein levels in seven retinal locations 2 weeks (three eyes) or 12 weeks (two eyes) after injection of 100 μl of PBAE NPs containing 38.4 μg of pCAG-GFP-Z1. Comparing these data to those in Fig. 5A, it appears that doubling the volume and dose of vector did not result in major differences in retinal GFP protein levels, but in two of three eyes, there were some locations of RPE/choroid with high levels (Fig. 5D). At this dose, GFP protein levels were similar at 2 and 12 weeks after injection for both retina and RPE/choroid. Compared with lower doses, there did not appear to be a major boost in GFP protein levels 2 weeks after suprachoroidal injection of 200 μl of PBAE NPs containing 76.8 μg of pCAG-GFP-Z1, but GFP levels were high at all locations in retina and RPE/choroid 12 weeks after injection (Fig. 5, E and F). The mean of all GFP levels at all retinal locations 2 weeks versus 12 weeks after injection of 200 μl of PBAE NPs was 37.0 versus 284.4 pg/mg protein (P = 0.07 by linear mixed-effects models), and the comparison for RPE/choroid was 96.7 versus 879.5 (P < 0.0001).
Fig. 5. Comparison of GFP expression at 2 and 12 weeks after suprachoroidal injection of 50, 100, or 200 μl containing pCAG-GFP-Z1.
Yucatan minipigs were given a suprachoroidal injection of 50, 100, or 200 μl of PBAE NP containing pCAG-GFP-Z1. At 2 weeks (three eyes for each volume) or 12 weeks (two eyes for each volume) after injection of 50 μl (A and B), 100 μl (C and D), or 200 μl (E and F) of PBAE NP, GFP protein was measured by ELISA in punches of retina [(A), (C), and (E)] or RPE/choroid [(B), (D), and (F)] at locations designated in key.
Effect of minimizing bacterial sequences and CpG repeats in expression plasmid
The presence of bacterial sequences, particularly unmethylated CpG motifs, in expression plasmids increases the likelihood of an immune response (11–15), and therefore, we contracted a commercial vendor (Aldevron, Fargo, ND) to minimize CpG motifs and other bacterial sequences in pCAG-GFP-Z1. This resulted in a 3322-bp plasmid (pCAG-GFP-nP). Two weeks after suprachoroidal injection of 50 μl of PBAE NPs containing 19.2 μg pCAG-GFP-nP, transverse sections anterior to the equator showed GFP fluorescence around the entire circumference of the eye (Fig. 6A) and high magnification showed GFP expression in cells of the inner retina as well as in photoreceptors (Fig. 6B). Immunohistochemical staining for GFP showed that the fluorescence in the inner and outer retina was due to expression of GFP and not autofluorescence (fig. S4). Sections far posterior to the equator also showed GFP fluorescence around the entire circumference of the eye (Fig. 6C), and high magnification showed GFP expression predominantly in photoreceptors but also in some cells of the inner retina (Fig. 6D). There was a trend toward higher GFP protein levels in retina and RPE/choroid, but due to substantial variability, the differences were not statistically significant (Fig. 6, E and F).
Fig. 6. Effect of minimizing CpG bacterial sequences in GFP expression plasmid.
Yucatan minipigs (n = 3) were given a suprachoroidal injection of 50 μl of PBAE NP containing 19.2 μg of pCAG-GFP-nP, in which bacterial sequences had been minimized. Two weeks after injection, one eye of each pig was used for localization of GFP expression in serial ocular frozen sections and the fellow eyes were used to measure GFP expression by ELISA in retina and RPE/choroid at seven locations. A representative ocular section anterior to the equator showed GFP fluorescence around the entire circumference of the eye (A), and high magnification showed GFP expression in cells of the inner retina as well as in photoreceptors (B). A section 7.49 mm posterior to the equator also showed GFP fluorescence around the entire circumference of the eye (C), and high magnification showed GFP fluorescence predominantly in photoreceptors, but also in some inner retinal cells (D). The mean level of GFP protein in the retina (E) or the RPE/choroid (F) calculated from seven measurements in each of three eyes was 219.2 and 649.8 pg/mg total protein after injection of pCAG-GFP-nP compared with 51.2 and 68.2 after injection of pCAG-GFP-Z1. Because of variability, differences were not statistically significant. *P = 0.26; **P = 0.08 by linear mixed-effects models.
Effect of multiple 50-μl suprachoroidal injections of PBAE NP containing 19.2 μg of pCAG-GFP-nP
Examination of GFP protein levels at each of the retinal locations sampled 2 weeks after a single superior injection of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-nP showed considerable regional variability within each of three eyes and between the eyes (Fig. 7A), and the same was true for RPE/choroid (Fig. 7B). To test whether nonuniform spread of PBAE NPs in the suprachoroidal space might contribute to this variability in GFP expression, three injections were given at different locations around the circumference of three eyes. One injection of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-nP was given superiorly, one was given temporally, and one was given inferiorly, all 4.5 mm posterior to the limbus with approximately 5 min in between injections. Two weeks after injections, there were high levels of GFP protein at all locations sampled in retina (Fig. 7C) and RPE/choroid (Fig. 7D), indicating a reduction in within-eye variability. Dot plots of all 21 measurements in the retinas (Fig. 7E) and RPE/choroid (Fig. 7F) of the three eyes given one injection versus those given three injections showed a marked reduction in the coefficient of variation for GFP protein levels in retina and RPE/choroid after three injections, indicating a substantial reduction in between-eye as well as within-eye variability. Mean GFP protein levels trended higher after three injections, but the differences were not statistically significant. No toxicity was observed in these eyes or any of the eyes treated with pCAG-GFP-Z1 NPs or pCAG-GFP-nP NPs at any of the dosages evaluated.
Fig. 7. Three suprachoroidal injections of PBAE NPs containing pCAG-GFP-nP at different locations reduce variability of GFP expression throughout retina and RPE/choroid.
The level of GFP protein (pg/mg total protein) measured at seven locations in the retina (A) and RPE/choroid (B) 2 weeks after a single suprachoroidal injection at 12:00 of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-nP in three eyes is compared with the level of GFP protein measured at the same seven locations in the retina (C) and RPE/choroid (D) 2 weeks after three injections (one superiorly, one temporally, and one inferiorly) of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-nP. Dot plots of all GFP protein levels at all locations in each eye show that compared with eyes given a single injection, those given a triple injection had a marked reduction in coefficient of variation (CV) of GFP levels in retina (E) and RPE/choroid (F).
DISCUSSION
There are many inherited retinal degenerations that cause substantial visual disability. Many of these degenerations are caused by loss-of-function mutations in genes critical for photoreceptor function and survival. Gene therapy to replace or augment the deficient protein has the potential to cure these blinding conditions if done at an early stage of disease. The most common strategy is subretinal injection of an AAV vector expressing the needed protein. The subretinal injection results in separation of photoreceptors from RPE by the vector-containing fluid, causing a small retinal detachment referred to as a bleb. This has the advantage of providing a high concentration of vector in close proximity to photoreceptors and RPE, resulting in good transfection efficiency and high expression of the therapeutic protein. However, a disadvantage of this approach is that there is little expression throughout the remainder of the retina, and unless the injection volume is large, resulting in a large bleb that involves a large area of retina, most of the retina remains untreated. This is a major problem because it means that the gene defect is not corrected in rods throughout a large area of the retina and those rods are destined for degeneration. When a sufficient number of rods degenerate, oxygen utilization is decreased and oxygen levels are increased (16, 17), resulting in oxidative damage to remaining rods and cones, causing gradual, progressive degeneration of the remaining retina (18–22). Therefore, gene replacement should be done early before there is extensive rod degeneration and should target as many rods as possible.
Another potential disadvantage is that separation of photoreceptors from RPE is potentially damaging to photoreceptors, and while it is possible for the photoreceptors to recover, that recovery may be incomplete so that a bleb involving the fovea can result in permanent reduction in vision. A third issue is that there is still uncertainty regarding the manner in which the immune system responds to transfection of cells by AAV. There may be initial inflammation after ocular gene therapy with an AAV vector that varies depending upon route of administration and is usually mild and manageable after subretinal injection, but what is more concerning are some signs of possible late immune response in some patients that may cause retinal damage. Some patients treated with subretinal injection of voretigene Neparvovec-rzyl, an AAV2 vector expressing RPE65, have developed progressive perifoveal chorioretinal atrophy (2), which could be an immune response to late presentation of viral antigens by transfected retinal neurons.
Suprachoroidal ocular gene therapy with PBAE polymeric NPs has the potential to address the above issues. We have previously shown that a 3-μl suprachoroidal injection of PBAE NPs containing 1 μg of a GFP expression plasmid in rats resulted in GFP expression in photoreceptors and adjacent RPE throughout the entire retina (9). Here, we have demonstrated that a 50-μl suprachoroidal injection of PBAE NPs containing 19.2 μg of GFP expression plasmid resulted in widespread GFP expression in photoreceptors and RPE in the minipig eye. Transverse sections extending as far posterior as possible showed detectable GFP expression in photoreceptors and RPE around the entire circumference of the eye even in the posterior-most sections. Quantitative analysis was done by measuring GFP protein levels by ELISA in retinal and RPE/choroid samples obtained from seven locations spanning the entire eyecup. This confirmed that after a single anterior suprachoroidal injection of PBAE NPs containing 19.2 μg of pCAG-GFP-Z1, GFP protein was detectable in all parts of the retina, but with considerable regional variability. There was also variability in RPE/choroid with GFP levels low at most locations in most eyes, but higher in some locations. In general, eyes that had higher GFP at some locations in RPE/choroid also tended to have high GFP at some locations in retina.
It was hypothesized that increasing the volume, which would also increase the amount of plasmid/transgene copies injected, would expand the suprachoroidal space to a greater degree and permit more uniform spread and more uniform and higher GFP expression. Two weeks after injection of higher volumes of NPs, there was still high within-eye and between-eye variability, making it difficult to assess the impact of increasing the dose. Expression of GFP was similar 2 and 12 weeks after injection of 50 or 100 μl of PBAE NPs containing 19.2 or 38.4 μg of pCAG-GFP-Z1, but there was a statistically significant increase in RPE/choroid and a trend toward an increase in retina between 2 and 12 weeks after injection of 200 μl of PBAE NPs containing 76.8 μg of pCAG-GFP-Z1. This suggests that peak expression after injection of higher doses of vector occurs at some point after 2 weeks.
It was tested whether reducing bacterial CpG sequences in the expression plasmid could increase expression of GFP. There was a trend toward higher GFP levels in retina and RPE/choroid 2 weeks after injection of PBAE NPs containing the modified plasmid (pCAG-GFP-nP) versus the parent plasmid, but differences were not statistically significant due to within-eye and between-eye variability. Although increasing the volume of injection failed to reduce variability in expression, it was felt that inconsistent spread of vector in the suprachoroidal space might still be responsible for the variability and might be overcome by performing multiple injections at different locations around the circumference of the eye. Suprachoroidal injections of 50 μl of PBAE NPs containing 19.2 μg of pCAG-GFP-nP were done superiorly, temporally, and inferiorly (inability to obtain good exposure nasally in the pig prevented injection on that side of the eye). Two weeks after this triple injection, there was a substantial reduction in within-eye and between-eye variability demonstrated by a marked reduction in the coefficient of variation compared with that seen after a single injection. This confirmed that nonuniform spread of vector in the suprachoroidal space is the cause of variable expression throughout the eye and between eyes given a single injection. Anatomic differences between eyes and technical differences regarding injections such as needle depth and orientation and rate of injection might contribute to nonuniform spread of vector in the suprachoroidal space.
No signs of toxicity were observed following administration of the biodegradable PBAE NPs in minipigs, suggesting that this approach may be a safe nonviral method for ocular gene transfer.
Compared with experiments in which a large number of rodents are given suprachoroidal injections at a single session providing large experimental numbers, suprachoroidal injections in a pig require general anesthesia and a large team of investigators, and a maximum of three pigs can be injected per session. This greatly reduces experimental numbers, making statistical comparisons difficult. Despite these challenges, the studies reported here provide a good indication of the optimal dose and suprachoroidal injection technique of PBAE NPs needed to obtain good reporter gene expression throughout the entire retina and RPE/choroid in a human size eye. This provides the foundation needed to begin expression studies with therapeutic transgenes in minipigs and provides additional evidence of the feasibility of using nonviral gene transfer for the treatment of inherited retinal degenerations.
MATERIALS AND METHODS
Experimental design
The study was designed to optimize the level and distribution of GFP expression in retina and RPE/choroid after suprachoroidal injection of PBAE NPs containing a GFP expression plasmid. Expression level was quantified by performing ELISA on samples collected throughout eyes. It was planned to test the effect of increasing plasmid dose by increasing the volume of injection and by doing repeated injections. Experiments were designed to assess the stability of expression over time between 2 and 12 weeks after injection and to assess the effect of reducing bacterial sequences in the expression plasmid.
Polymer synthesis
1,4-Butanediol diacrylate (B4), 5-amino-1-pentanol (S5), and 1-(3-aminopropyl)-4-methyl-piperazine (E7) were purchased from Alfa Aesar (Ward Hill, MA). PBAE polymer was synthesized by a two-step reaction. First, acrylate-terminated base polymer (B4S5) was synthesized by Michael addition reaction of B4 with S5 at 1.1:1 acrylate:amine monomer molar ratio in the dark under magnetic stirring for 24 hours at 90°C. In the second step, the acrylate-terminated base polymer was end-capped through another Michael addition reaction in the presence of excess of primary amine-containing small-molecule E7. Briefly, the polymer (200 mg/ml) was mixed with 0.5 M E7 in anhydrous tetrahydrofuran (THF) at room temperature for 2 hours. The final polymer (B4S5E7, or 457) was purified by precipitation into diethyl ether and stored in anhydrous dimethyl sulfoxide (DMSO) at 100 mg/ml with desiccant at −20°C until use. Polymer molecular weight was assessed via gel permeation chromatography (Agilent, Savage, MD) relative to polystyrene standards, and the number average and weight average molecular weight were 8610 and 43,500 g/mol, respectively.
Plasmid preparation
pEGFP-N1 was obtained from Takara Bio USA Inc. (Mountain View, CA). pCAG-GFP-Z1 was prepared to minimize total plasmid length and eliminate unwanted sequences in the parental plasmid constructs. A linear DNA fragment containing a multiple cloning site, bacterial zeocin resistance gene, and bacterial origin of replication was first synthesized by Twist Bioscience (South San Francisco, CA) with Eco RI sites at the 5′ and 3′ ends (1397 bp). The linear DNA fragment was then digested with Eco RI and ligated to form an empty plasmid containing the multiple cloning sites, bacterial zeocin resistance gene, and pUC bacterial origin of replication (1391 bp). The SV40 polyadenylation sequence was introduced from pUNO1-m41BBL (InvivoGen, San Diego, CA) using 5′/3′ enzyme pair of Nhe I and Eco RI. The CAG promoter sequence was introduced from pPB-CAG-GFPd2 (Addgene 115665) using Spe I and Kpn I 5′ and 3′ restriction enzyme cloning (1726 bp). Enhanced GFP (eGFP) was then introduced from eGFP-N1 using restriction enzyme pair 5′/3′ of Age I and Xba I. All restriction enzymes were purchased from New England Biolabs (Ipswich, MA). A GFP expression plasmid with most bacterial and CpG sequences removed, pCAG-GFP-nP, was obtained from Aldevron (Fargo, ND).
NP formulation
pDNA-carrying NPs were formulated by electrostatic binding of positively charged PBAE polymer and negatively charged expression plasmids (pCAG-GFP-Z1 and pCAG-GFP-nP) as previously described (9). Briefly, 457 PBAE polymer in DMSO at 100 mg/ml and pDNA in water were both diluted with 25 mM sodium acetate pH 5 to 5.05 and 0.31 mg/ml, respectively. Then, polymer and pDNA solutions were mixed at 3:2 (v/v) ratio for 25 (w/w) ratio of polymer to DNA and incubated for 10 min to allow particle complexation. To lyophilize the NPs, the final NP solution was mixed with sucrose as a cryoprotectant to a final concentration of 30 mg/ml, then aliquoted, and lyophilized. Lyophilized NPs were stored with desiccant at −80°C until use. Immediately before injection, lyophilized NPs were reconstituted with sterile water to a final sucrose concentration of 100 mg/ml.
NP characterization
Lyophilized NPs were resuspended in sterile water to a final concentration of 0.38 μg/μl, and 20 μl of NPs was diluted in 1 ml of 0.1× phosphate-buffered saline (PBS). Particle size (hydrodynamic diameter) was assessed by DLS using Malvern Zetasizer Pro (Malvern Panalytical, Malvern, UK). Surface charge (zeta potential) was assessed by electrophoretic mobility using Malvern Zetasizer Pro. Size and surface charge were assessed on n = 3 individually prepared NP replicates. The number-weighted size distribution was also measured by NTA using NanoSight NS300 (Malvern Panalytical) after diluting each of three individually prepared batches of particles 500-fold in 1× PBS. The number of plasmids per particle was calculated by NTA as has been previously described (23).
To evaluate the morphology and to confirm the size of the NPs, samples were imaged with a Hitachi 7600 TEM (Hitachi High-Tech, Tokyo, Japan). Lyophilized particles were resuspended in water to a DNA concentration of 0.38 μg/μl. Samples were diluted further to a concentration of 0.001 μg/μl, transferred to a carbon film 400-mesh copper grid (Electron Microscopy Sciences, Hatfield, PA), and allowed to dry for 4 hours. Following this period, 1% uranyl acetate solution (Electron Microscopy Sciences, Hatfield, PA) was added to the copper grid. The grids were then washed with deionized water, dried overnight, and imaged via TEM.
To assess NP DNA binding, DNA NPs and naked DNA were loaded into a 1% agarose gel with ethidium bromide (1 μg/ml; 250 ng per lane) and imaged following electrophoresis.
Experimental animals
Ten 3- to 4-month-old Göttingen minipigs (Marshall Bio Resources, North Rose, NY) and fifteen 3- to 4-month-old Yucatan minipigs (Sinclair Bio Resources, Auxvasse, MO) were treated in accordance with the Association for Research in Vision and Ophthalmology Statement for Use of Animals in Ophthalmic and Vision Research, and protocols were reviewed and approved by the Johns Hopkins University Animal Care and Use Committee. Minipigs were anesthetized using a combination of (i) ketamine hydrochloride (11 to 33 mg/kg, intramuscularly) for sedation; (ii) butorphanol 0.2 mg/kg + midazolam 0.5 mg/kg + acepromazine 0.2 mg/kg combined, intramuscularly; (iii) ketamine 20 to 30 mg/kg + xylazine 2 to 3 mg/kg intramuscularly followed by reversal with yohimbine 0.11 mg/kg intramuscularly or intravenously. In all cases, pigs were fasted overnight before anesthesia. Anesthesia was maintained by inhalation of 0.5% to 2.5% isoflurane and 100% O2. During anesthesia, heart rate and O2 saturation were monitored by pulse oximetry. After sedation, the eyes were sterilized with a drop of povidone-iodine solution USP 10% (Ricca Chemical Company, Arlington, TX) and numbed with a drop of proparacaine hydrochloride ophthalmic solution USP 0.5% (Bausch & Lomb, Bridgewater, NJ). Fundus images were taken before and after suprachoroidal injections. Pupils were dilated with 1% tropicamide (Alcon Labs Inc., Fort Worth, TX) and phenylephrine hydrochloride 2.5% (Paragon BioTeck, Portland, OR), and both eyes received a drop of ophthalmic hypromellose solution GenTeal (Alcon, Fort Worth, TX). The lens of a RetCam3 fundus camera (Natus, Middleton, WI) was gently touched to the cornea and used to photograph the fundus before and after injections. After suprachoroidal injection, a small amount of triple antibiotic ointment bacitracin (First Aid Research Corp., Jupiter, FL) was applied to lubricate the eye until the pigs wake up and reduce the minimal chance of infection. At the end of the experiment, euthanasia was performed by giving an overdose of pentobarbital (150 mg/kg intravenously—Euthasol 390 mg/ml, Virbac, Westlake, TX).
Suprachoroidal injection of NPs
Injections were done with sterile instruments under sterile conditions. After minipigs were anesthetized, topical 0.5% proparacaine and 10% povidone iodine eye drops were administered (anesthesia was detailed in the previous section). A sterile speculum was placed to hold the eyelids open. The retina was visualized with a handheld RetCam fundus camera, and pre-injection fundus photographs were taken. Eyes were visualized with a surgical microscope (Zeiss, Oberkochen, Germany), and the conjunctiva was dissected to expose the sclera near the limbus. A 30-gauge needle on a 1-ml syringe was used to make an oblique, partial-thickness scleral tunnel 4.5 mm posterior to the limbus, and then a blunt-tip needle attached to a Hamilton syringe (Hamilton, Reno, NV) was used to enter the suprachoroidal space. A 100-μl Hamilton #710 syringe with an attached 34-gauge 45° beveled blunt needle was used for 50- or 100-μl injections, and a Hamilton #1725 syringe and 33-gauge/11-mm needle (PRE-33013 Acuderm Inc., FL) was used for 200-μl injections. Ocular massage was used to reduce intraocular pressure before 200 μl injections, and the injections were done very slowly. After injection, a cotton swab was held over the injection site for about 60 s before removal of the needle. Post-injection fundus photographs were taken with the RetCam, and the entire retina was examined with indirect ophthalmoscopy.
Histology and immunohistochemistry
After enucleation, minipig eyes were fixed in 10% formalin at room temperature for 4 to 6 hours, the cornea was removed, and the eye was filled with 10% formalin and fixed overnight at 4°C. The anterior segment was removed under a dissecting microscope, and the vitreous cavity was flushed with 10% formalin using a 23-gauge needle to penetrate into the vitreous with the fixative. The eyecup was then incubated at 4°C in 10% formalin for 12 hours, followed by PBS containing 15% sucrose for 12 hours, and then PBS containing 30% sucrose for 12 hours. Under a dissecting microscope, PBS was injected through a 20-gauge needle to gently dissect and remove the remaining vitreous and the eyecup was filled with optimal cutting temperature (OCT) embedding medium and placed in a −80°C freezer overnight. Thirty-five-micrometer transverse sections were cut starting at the anterior edge of the eyecup and proceeding posteriorly.
Slides were dried and examined with a Zeiss fluorescence microscope. Because of the large size of the sections, the entire circumference of the eye could not be obtained in a single image. Therefore, overlapping images were taken around the entire circumference and merged using Photoshop Photomerge function or ImageJ (https://imagej.nih.gov/ij/download.html). Some sections were immunohistochemically stained for GFP. Nonspecific binding was blocked by a 30-min incubation in 8% normal rabbit serum at 25°C. The sections were incubated with a rabbit polyclonal antibody (1:300) against eGFP conjugated with Alexa Fluor 594 (A-21312, Thermo Fisher Scientific, Waltham, MA) at 23°C for 2 hours. After washing with PBS containing 0.05% Tween 20, slides were counterstained with Hoechst 33258 (861405, Sigma, St. Louis, MO) and examined by fluorescence microscopy. Some ocular sections were stained with hematoxylin and eosin.
Measurement of GFP protein levels by ELISA
After enucleation, eyes were kept on ice, and under a dissecting microscope, the anterior segment and vitreous were removed. A 7-mm corneal trephine was used to obtain circular punches of retina at seven locations (fig. S4). The trephine was then used to dissect deeper and obtain circular RPE/choroid samples from the same seven locations. The retina and choroid/RPE samples were placed in 300 μl of PBS containing protease inhibitor cocktail (11836170001, Roche, Mannheim, Germany) and sonicated for 5 s. Samples were centrifuged at 14,000 rpm for 15 min, and the protein concentrations were determined by a Bio-Rad Protein Assay Dye Reagent Concentrate (#5000006, Bio-Rad, Hercules, CA), using bovine serum albumin (BSA) as the standard. GFP protein levels were measured using a GFP SimpleStep ELISA kit (ab171581, Abcam, Cambridge, MA). Briefly, 50 μl of sample or GFP standard dilutions was added to duplicate wells of 96-well plates, followed by 50 μl of GFP capture antibody and GFP detect antibody mixture. Plates were incubated at 23°C for 1 hour and washed five times with rinse buffer, and after addition of 100 μl of 3,3′,5,5′-tetramethylbenzidine (TMB) substrate solution, they were incubated at 23°C in the dark for 10 min. After addition of 100 μl of stop solution, absorption was measured at 450 nm with the SpectraMax Plus 384 Microplate Reader.
Statistical analysis
Statistical comparisons for treatment effects in each experiment were determined using general linear mixed models, and graphics were done using GraphPad Prism software v. 5.0. All measurements from either eye of a pig were assumed to be exchangeable when modeling correlation structure and were assumed to be subject to non-error variability. A P value less than 0.05 was considered significant.
Acknowledgments
Funding: This work was supported by R01EY031097 (P.A.C. and J.J.G.), R01EY017549 (J.J.G and P.A.C.), and the Biostatistics Module of P30EY001765 from the National Eye Institute. It was also supported by grants from Cove Therapeutics Inc. (P.A.C. and J.J.G.), the Barth Foundation (P.A.C.), Fighting Blindness Canada (P.A.C.), the Gosnell Foundation (P.A.C.), and Research to Prevent Blindness (P.A.C. and J.J.G.), and gifts from Per Bang-Jensen (P.A.C.) and Andrew and Yvette Marriott (P.A.C.).
Author contributions: J.S. helped to design experiments, performed experiments, analyzed data, and edited and approved the manuscript. R.L.e.S. helped to design experiments, performed experiments, analyzed data, prepared figures, and edited and approved the manuscript. M.Z. helped to design experiments, performed experiments, analyzed data, and edited and approved the manuscript. K.M.L. helped to design experiments, performed experiments, prepared figures, and edited and approved the manuscript. S.F.H. performed experiments, analyzed data, and edited and approved the manuscript. S.Y.T. helped to design experiments, performed experiments, analyzed data, and edited and approved the manuscript. S.M.L. performed experiments, and edited and approved the manuscript. S.R.S. performed experiments and edited and approved the manuscript. D.R.W. performed experiments and edited and approved the manuscript. J.J.G. developed PBAE vectors, designed experiments, analyzed data, and edited and approved the manuscript. P.A.C. designed experiments, analyzed data, wrote the first draft of the manuscript, and approved the manuscript.
Competing interests: These arrangements have been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies. J.S.: Related to the current manuscript—Patent applications on suprachoroidal nonviral gene transfer. J.J.G.: Related to the current manuscript—Patent applications on suprachoroidal nonviral gene transfer and PBAE NPs; Cove Therapeutics Inc.: cofounder, member of Board of Directors, officer, and equity. Unrelated to current manuscript—ASCLEPIX THERAPEUTICS: Board of Directors member and consultant, equity; DOME THERAPEUTICS: managing member, officer, equity; VASORX: Board of Directors member, equity; ONCOSWITCH: manager, equity; WYVERNA: manager, equity; P.A.C.: Related to current the manuscript—Patent applications on suprachoroidal nonviral gene transfer; Cove Therapeutics Inc.: consultant, equity. Unrelated to the current manuscript—ALLEGRO: advisory board, equity; ASHVATTHA THERAPEUTICS: consultant; BAUSCH and LOMB: consultant; CATAWBA RESEARCH: consultant; CELANESE: consultant; grant; CLEARSIDE: consultant; Codiak: consultant; EXGENESISBIO: consultant; EXONATE LTD: consultant; GENENTECH/ROCHE INC: advisory board, honoraria to JHU, investigator, grants; SANOFI GENZYME: investigator, grant; GRAYBUG VISION: cofounder, equity; MALLINCKRODT PHARMACEUTICALS: grant; MERCK & CO, INC: consultant; OXFORD BIOMEDICA: investigator, grant; PERFUSE; consultant; REGENXBIO, INC: investigator, grant.
Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. The nanoparticles can be provided by Johns Hopkins University pending scientific review and a completed material transfer agreement. Requests for the nanoparticles should be submitted to: green@jhu.edu.
Supplementary Materials
This PDF file includes:
Figs. S1 to S4
REFERENCES AND NOTES
- 1.Russell S., Bennett J., Wellman J. A., Chung D. C., Yu Z.-F., Tillman A., Wittes J., Pappas J., Elci O., McCague S., Cross D., Marshall K. A., Walshire J., Kehoe T. L., Reichert H., Davis M. D., Raffini L., George L. A., Hudson F. P., Dingfield L., Zhu X., Haller J. A., Sohn E. H., Mahahan V. B., Pfeifer W., Weckmann M., Johnson C., Gewaily D., Drack A., Stone E. M., Wachtel K., Simonelli F., Leroy B. P., Wright J. F., High K. A., Maguire A. M., Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: A randomised, controlled, open-label, phase 3 trial. Lancet 390, 849–860 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gange W. S., Sisk R. A., Besirli C. G., Lee T. C., Havunjian M., Schwartz H., Borchert M., Sengillo J. D., Mendoza C., Berrocal A. M., Nagiel A., Perifoveal chorioretinal atrophy after subretinal voretigene Neparvovec-rzyl for RPE65-mediated leber congenital amaurosis. Ophthamol. Retina 6, 58–64 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Green J. J., Langer R., Anderson D. G., A combinatorial polymer library approach yields insight into nonviral gene delivery. Acc. Chem. Res. 41, 749–759 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sunshine J. C., Peng D. Y., Green J. J., Uptake and transfection with polymeric nanoparticles are dependent on polymer end-group structure, but largely independent of nanoparticle physical and chemical properties. Mol. Pharm. 9, 3375–3383 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Guerrero-Cazares H., Tzeng S. Y., Young N. P., Abutaleb A. O., Quinones-Hinojosa A., Green J. J., Biodegradable polymeric nanoparticles show high efficacy and specificity at DNA delivery to human glioblastoma in vitro and in vivo. ACS Nano 8, 5141–5153 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rui Y., Wilson D. R., Tzeng S. Y., Yamagata H., Sudhakar D., Conge M., Berlinicke C. A., Zack D. J., Tuesca A., Green J. J., High-throughput and high-content bioassay enables tuning of polyester nanoparticles for cellular uptake, ednosomalescape, andsystemic in vivodelivery of mRNA. Sci. Adv. 8, eabk2855 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Herrera-Barrera M., Ryals R. C., Gautam M., Jozic A., Landry M., Korzun T., Gupta M., Acosta C., Stoddard J., Reynaga R., Tschetter W., Jacomino N., Taratula O., Sun C., Lauer A. K., Neuringer M., Sahay G., Peptide-guided lipid nanoparticles deliver mRNA to the neural retina of rodents and nonhuman primates. Sci. Adv. 9, eadd4623 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Drag S., Dotiwala F., Upadhyay A. K., Gene therapy for retinal degenerative diseases: Progress, challenges, and future directions. Invest. Ophthalmol. Vis. Sci. 64, 39 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shen J., Kim J., Tzeng S. Y., Ding K., Hafiz Z., L. D., J. Wang, J. J. Green, P. A. Campochiaro, Suprachoroidal gene transfer with nonviral nanoparticles. Sci. Adv. 6, eaba1606 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vrolyk V., Desmariais M.-J., Lambert D., Haruna J., Benoit-Biancamano M., Neonatal and juvenile ocular development in Göttingen minipigs and domestic pigs: A histomorphological and immunohistochemical study. Vet. Pathol. 57, 889–914 (2020). [DOI] [PubMed] [Google Scholar]
- 11.Hacker G., Redecke V., Hacker H., Activation of the immune system by bacterial CpG-DNA. Immunology 105, 245–251 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kinman D. M., Takeshita F., Fursel I., Leifer C., Ishii K. J., Verthelyi D., Gursel M., CpG DNA: Recognition by and activation of monocytes. Micobes Infect. 4, 897–901 (2002). [DOI] [PubMed] [Google Scholar]
- 13.Reyes-Sandoval A., Ertl H. C. J., CpG methylation of a plasmid vector results in extended transgene product expression by Circumventing induction of immune responses. Mol. Ther. 9, 249–261 (2004). [DOI] [PubMed] [Google Scholar]
- 14.Talati A. J., Kim H. J., Kim Y. I., Yi A. K., English B. K., Role of bacterial DNA in macrophage activation by group B streptococci. Microbes Infect. 10, 1106–1113 (2008). [DOI] [PubMed] [Google Scholar]
- 15.Hyde S. C., Pringle I. A., Abdullah S., Lawton A. E., Davies A., Varathalingam A., Nunez-Alonso G., Green A., Bazzani R. P., Sumner-Jones S. G., Chan M., Li H., Yew N. S., Cheng S. H., Boyd A. C., Davies J. C., Griesenbach U., Porteous D. J., Sheppard D. N., Munkonge F. M., Alton E. W. F. W., Gill D. R., CpG-free plasmids confer reduced inflammation and sustained pulmonary gene expression. Nat. Biotechnol. 26, 549–551 (2008). [DOI] [PubMed] [Google Scholar]
- 16.Yu D. Y., Cringle S. J., Su E. N., Yu P. K., Intraretinal oxygen levels before and after photoreceptor loss in the RCS rat. Invest. Ophthalmol. Vis. Sci. 41, 3999–4006 (2000). [PubMed] [Google Scholar]
- 17.Yu D. Y., Cringle S. J., Valter K., Walsh N., Lee D., Stone J., Photoreceptor death, trophic factor expression, retinal oxygen status, and photoreceptor function in the P23H rat. Invest. Ophthalmol. Vis. Sci. 45, 2013–2019 (2004). [DOI] [PubMed] [Google Scholar]
- 18.Shen J., Yan X., Dong A., Petters R. M., Peng Y.-W., Wong F., Campochiaro P. A., Oxidative damage is a potential cause of cone cell death in retinitis pigmentosa. J. Cell. Physiol. 203, 457–464 (2005). [DOI] [PubMed] [Google Scholar]
- 19.Komeima K., Rogers B. S., Lu L., Campochiaro P. A., Antioxidants reduce cone cell death in a model of retinitis pigmentosa. Proc. Natl. Acad. Sci. U.S.A. 103, 11300–11305 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Komeima K., Rogers B. S., Campochiaro P. A., Antioxidants slow photoreceptor cell death in mouse models of retinitis pigmentosa. J. Cell. Physiol. 213, 809–815 (2007). [DOI] [PubMed] [Google Scholar]
- 21.Cideciyan A. V., Jacobson S. G., Beltran W. A., Sumaroka A., Swider M., Iwabe S., Roman A. J., Olivares M. B., Schwartz S. B., Komaromy A. M., Hauswirth W. W., Aguirre G. D., Human retinal gene therapy for Leber congenital amaurosis shows advancing retinal degeneration despite enduring visual improvement. Proc. Natl. Acad. Sci. U.S.A. 110, E517–E525 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jacobson S. G., Cideciyan A. V., Roman A. J., Sumaroka A., Schwartz S. B., Heon E., Hauswirth W. W., Improvement and decline in vision with gene therapy in childhood blindness. N. Engl. J. Med. 372, 1920–1926 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bhise N. S., Shmueli R. B., Gonzalez J., Green J. J., A novel assay for quantifying the number of plasmids encapsulated by polymer nanoparticles. Small 8, 367–373 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figs. S1 to S4