Introduction of DNA Into the Rat and Primate Trabecular Meshwork by Fusogenic Liposomes

Introduction of DNA into the Rat and Primate Trabecular Meshwork by Fusogenic Liposomes

Masanori Hangai,1 Hidenobu Tanihara,l Yoshihito Honda,1 and Yasufumi Kaneda'

PURPOSE. TO evaluate the feasibility of introducing exogenous genes and phosphorothioate oligonucleotides into the anterior chamber tissues of rats and monkeys using the authors' fusogenic liposomes. METHODS. Hemagglutinating virus of Japan liposomes containing LacZ DNA-high-mobility group 1 complexes or fluorescein isothiocyanate (FITC)-labeled phosphorothioate oligonucleotides were prepared and injected into the anterior chambers of rats (3 /xl) and rhesus monkeys (30 /u-i). The expression of LacZ DNA was visualized histochemically by j3-Galactosidase assay and was followed for as long as 60 days in rats and 30 days in monkeys. FITC-labeled phosphorothioate oligonucleotides were observed by fluorescence microscopy for as long as 14 days in rats and 7 days in monkeys. RESULTS. Injection of LacZ DNA-high-mobility group 1 complexes encapsulated in hemagglutinating virus of Japan liposomes resulted in blue staining in the trabecular meshwork and iris-ciliary body of rats and selectively in the trabecular meshwork of monkeys at the concentrations used. This LacZ expression lasted for as long as 14 days after injection in both animals. Phosphorothioate oligonucleotides (3 /xM) also were introduced into the rat trabecular meshwork and iris-ciliary body and into the primate trabecular meshwork when encapsulated in hemagglutinating virus of Japan liposomes, although the injection of naked FITC-labeled phosphorothioate oligonucleotides at the same concentration resulted in little fluorescence in any anterior chamber tissue.

CONCLUSIONS. This study shows that the use of hemagglutinating virus of Japan liposomes can transfer LacZ DNA and phosphorothioate oligonucleotides to adult rat and primate trabecular meshwork. This system may enable progress in glaucoma research and in the development of nonviral somatic gene therapy of the trabecular meshwork to treat glaucoma. (Lnvest Ophthalmol Vis Set. 1998;39:509-5l6)

laucoma is the second leading cause of vision loss in of the trabecular meshwork of eyes from patients with POAG the world; it affects at least 1 in 100 people world- and with glucocorticoid-induced glaucoma.6 The trabecular wide. ' Glaucoma is generally associated with elevated meshwork cells and extracellular matrix components also are G 2 3 intraocular pressure, ' which has been shown to lead to the thought to play a crucial role in the normal resistance to apoptotic death of retinal ganglion cells in experimental glau- aqueous outflow.6'7 It is hypothesized that a disturbed metab- coma.4'5 Therapeutic efforts have been directed largely at in- olism of the trabecular extracellular matrix may be the major traocular pressure reduction, based on the consensus that cause of increased outflow resistance in glaucoma. To make reduced intraocular pressure decreases the risk for the progres- clear the precise role of the extracellular matrix components in sion of visual loss.3 It is generally accepted that elevated in- outflow resistance, depletion, overexpression, or reconstitu- traocular pressure in most common types of glaucoma, such as tion of individual genes for trabecular extracellular matrix primary open-angle glaucoma (POAG), glucocorticoid-induced molecules, their receptors or extracellular matrix modulators, glaucoma, and pseudoexfoliation glaucoma, results from an such as transforming growth factor-/3 and matrix metallopro- increased resistance to aqueous outflow in the trabecular teinases, are necessary.6 This approach could also provide meshwork. Excessive, abnormal accumulations of extracellular evidence for altering the trabecular meshwork to treat glau- matrix materials, such as glycosaminoglycans and elastin, have coma. been noted in the corneoscleral and juxtacanalicular portions It has been recognized that POAG is hereditary, although not necessarily according to the simple Mendelian patterns. From the genetic linkage analysis of patients with POAG and a From the "Department of Ophthalmology and Visual Science, 8 12 Graduate School of Medicine, Kyoto University, Japan, and the insti- strong family history linked to chromosome lq, " several tute for Molecular and Cellular Biology, Osaka University, Japan. candidate genes have been tested, and mutations were recently Supported in part by a Grant-in-Aid for Scientific Research from detected in the gene encoding a trabecular meshwork-induced the Ministry of Education, Science, Sports, and Culture, by a Grant-in- glucocorticoid response protein.13 This protein is known to be Aid for Cancer Research from the Ministry of Health and Welfare of the Japanese Government, and by the Japanese National Society for the expressed by the trabecular meshwork and iris-ciliary body. Prevention of Blindness. Mutations in this gene were found in patients with a strong Submitted for publication April 21, 1997; revised September 16, family history and in unselected patients with POAG. To study 1997; accepted November 18, 1997. the mechanism by which the genetic abnormalities lead to Proprietary interest category: N. POAG, overexpression or inhibition of the gene products in Reprint requests: Masanori Hangai at present address: Doheny Eye Institute, University of Southern California School of Medicine, Room the trabecular meshwork in vivo would be useful. This could 303, 1450 San Pablo Street, Los Angeles, CA CA9OO33. also lead to somatic gene therapy.

Downloaded from iovs.arvojournals.org on 09/23/2021 510 Hangai et al. IOVS, March 1998, Vol. 39, No. 3

For these reasons, the introduction of genes or anti- sonicated for 3 seconds and was agitated by vortexing for 30 sense oligomers into trabecular meshwork cells in vivo may seconds. After the addition of 300 jal balanced salt solution, provide an opportunity to study molecular mechanisms of the sample was shaken at 37°C for 30 minutes. Hemagglu- glaucoma and to alter the trabecular meshwork as a treat- tinating virus of Japan (Z strain) was propagated, collected, ment for glaucoma. This approach would depend on target- and purified as described previously.21 The purified hemag- ing the human trabecular meshwork cells. Previous reports glutinating virus of Japan was inactivated by ultraviolet irra- showed that adenoviral vectors could introduce a reporter diation (110 erg/mm2 per second) for 3 minutes just before gene into the murine and rabbit trabecular meshwork in use. The liposome suspension (0.5 ml, containing 10 mg l4 16 vivo. ~ However, the human drainage system through the lipids) was mixed with the inactivated hemagglutinating trabecular meshwork is markedly different in architecture virus of Japan (30,000 hemagglutinating units) in a total from that of other species, except for other higher pri- volume of 4 ml balanced salt solution. The mixture was 718 mates.' Therefore, it is not necessarily appropriate to incubated for 5 minutes at 4°C and was shaken gently for 30 apply results obtained in lower mammals, such as rodents minutes at 37°C. Free hemagglutinating virus of Japan was and rabbits, to humans. The rhesus monkey is considered a removed from the hemagglutinating virus of Japan lipo- useful animal model for investigating the trabecular mesh- somes by sucrose density gradient centrifugation. The hem- work for glaucoma. The morphology, functions, and extra- agglutinating virus of Japan liposome solution was concen- cellular matrix synthesis of the monkey trabecular mesh- trated 8-fold by centrifugation at 27,000g for 30 minutes. work have been well documented. Therefore, we used not only rats but also rhesus monkeys as models for evaluating All animal experiments adhered to the ARVO Statement the ability of nonviral vectors, hemagglutinating virus of for the Use of Animals in Ophthalmic and Vision Research. Japan liposomes, to transfer LacZ DNA and phosphorothio- Sprague-Dawley rats (6 weeks old, male) and adult rhesus ate oligonucleotides into the trabecular meshwork in vivo. monkeys were anesthetized with an intramuscular injection of We show here that, using this system, LacZ DNA and phos- ketamine-HCl. Sixty-six rats were used for gene transfer exper- phorothioate oligonucleotides can be introduced into the iments and 21 for the introduction of oligomers. Twelve mon- rat and nonhuman primate trabecular meshwork cells. keys were used for gene transfer experiments, and another 12 were used for oligomer introduction. All injections were mon- itored under an ophthalmic operating microscope. Anterior chamber injections of hemagglutinating virus of Japan lipo- MATERIALS AND METHODS somes were made through the peripheral cornea. The amount pAct-lacZ (7.2 kb) carrying the gene for /3-Galactosidase (j3-Gal) of plasmid DNA injected was 0.24 jag (3 M>1) per rat and 2.4 jag of Escherichia colt under the control of a chicken /3-actin (30 /xl) per monkey. The amount of phosphorothioate oligo- promoter was constructed previously.l9 The plasmid DNA was nucleotides injected was 0.18 /xg (3 jal) per rat and 1.8 jag (30 purified by equilibrium centrifugation in cesium chloride con- jal) per monkey. Right eyes were used for each transfection, taining ethidium bromide and then was used for injections in and left ones were uninjected to serve as internal controls. naked forms or for preparations of hemagglutinating virus of Three preparations of hemagglutinating virus of Japan lipo- Japan liposomes. somes were used throughout this study. Scramble phosphorothioate oligonucleotides (5'-CT- Animals were killed by an overdose of pentobarbital. Eyes trans- TCGTCGGTACCGTCTT-3') were synthesized and labeled with fected with LacZ DNA were enucleated and fixed in 0.1% glutaral- fluorescein isothiocyanate (FITC) on the 3' and 5' ends. Briefly, dehyde and 2% formaldehyde in phosphate-buffered saline for 1 to 3 3',5'-di-amino-modified phosphorothioate oligonucleotides hours at 4°C. For /3-Gal assay, the fixedeye s were incubated for 12 were synthesized with phosphoamidite chemistry by using hours at 37°C with 1 mg/ml X-Gal (5-bromo-4-diloro-3-indolyl-/3-i> amino modifier controlled-pore glass and Beaucage reagents. galactopyranoside; Sigma Chemical, St. Louis, MO) in phosphate- Fluorescein was introduced into this oligomer, according to buffered saline containing 10 mM K3Fe(CN)6, 10 mM K4Fe(CN)6, the method previously described,20 and then was purified by a and 2 mM MgC12. After postfixation in 2.5% glutaraldehyde and 1% reversed-phase high-performance liquid chromatography procedure or by a nucleic acid purification column (NAP 10; formaldehyde in a 0.1 M sodium cacodylate buffer (pH 7.0) for 12 Pharmacia Fine Chemicals, Piscataway, NT). hours at 4°C, the eyes were frozen in OCT compound and were Hemagglutinating virus of Japan liposomes were pre- cryosectioned at a l4-/um thickness. pared as described previously.19'21'22 Briefly, high-mobility Eyes transfected with phosphorothioate oligonucleotides group 1 was purified from calf thymus, and DNA-high- were enucleated and fixed in 2.5% glutaraldehyde and 1% mobility group 1 complexes (200:64 pig) were formed by formaldehyde in a 0.1 M sodium cacodylate buffer (pH 7.0) for incubation for 30 minutes at 20°C.'9'21 Then, 10 mg of a 12 hours at 4°C. After cryosectioning at 14 jam, the sections dried lipid film consisting of phosphatidylserine, phosphati- were examined by fluorescence microscopy. Some sections dylcholine, and cholesterol (1:4.8:2 in weight ratio) was were stained with 5% propidium iodide (Oncor, Gaithersburg, formed on the sides of a flask by the removal of tetrahydro- MD) for 2 minutes to determine the sites of the nuclei. Other furan in a rotary evaporator. The lipid film was hydrated in sections were counterstained with 0.3% erichrome black T 200 jal balanced salt solution (137 mM NaCl, 5.4 raM KC1, 10 (Sigma) to eliminate possible autofluorescence of the elastic mM Tris-HCl, pH 7.6) containing the DNA-high-mobility fibers in the trabecular meshwork.23 group 1 complexes or FITC-labeled phosphorothioate oligo- We observed two quadrants (upper and lower) of rat eyes nucleotides (150 jag). The mixture was agitated by vortexing and all of the quadrants of monkey eyes. The intensity of blue for 30 seconds and then was left to stand for 30 seconds. staining and fluorescence was graded as heavy (3+), moderate This procedure was repeated nine times. The mixture was (2+), weak (1+), very weak (±), or absent (—).

Downloaded from iovs.arvojournals.org on 09/23/2021 FOVS, March 1998, Vol. 39, No. 3 Introduction of DNA into Trabecular Meshwork 511

SC TM

FIGURE 1. Hemagglutinating virus of Japan liposome-mediated introduction oiLacZ DNA and phosphorothioate oligonucleotides into rats. LacZ expression in the trabecular meshwork (TM) (a) and in the iris-ciliary body (b) 10 days after transfection. (c) Control eyes injected with naked LacZ DNA. (d-f) Introduction of fluorescein isothiocyanate (FITC)-labeled phosphorothioate oligonucleotides. Fluorescence was seen in the trabecular meshwork (d) and in the iris-ciliary body (e) of eyes introduced by hemagglutinating virus of Japan liposomes 1 hour after injection. Little fluorescence was seen in eyes injected with naked FITC-labeled phosphorothioate oligonucleotides. Co, cornea; SC, Schlemm's canal; I, iris; and CB, ciliary body. Magnification: X200 (a-c); X250 (d); and X125 (e,f).

RESULTS the rat anterior chambers has resulted in LacZ expression in the trabecular meshwork and iris-ciliary body (Figs, la, lb) but Introduction into the Rat Anterior Chamber not in the cornea or lens (data not shown). Blue staining Tissue seemed to be localized in the iris epithelium and ciliary vessels. The injection of LacZ DNA-high-mobility group 1 complexes LacZ expression mediated by hemagglutinating virus of Japan encapsulated in hemagglutinating virus of Japan liposomes into liposomes and high-mobility group 1 was most intense 7 days

Downloaded from iovs.arvojournals.org on 09/23/2021 512 Hangai et al. IOVS, March 1998, Vol. 39, No. 3

TABLE 1. Time Course of LacZ Expression in Rats and these organs, such as the aorta, showed autofluorescence in Monkeys normal rats, no fluorescence besides the autofluorescence was observed in rats injected with FITC-labeled phosphorothioate Rats Monkeys oligonucleotides encapsulated in hemagglutinating vims of Ja- pan liposomes. Number of Number of Day Eyes Intensity Eyes Intensity Introduction into the Primate Trabecular Meshwork i—i 0/3 _ 4 3/3 1 + 1/1 ± The injection of LacZ DNA-high-mobility group 1 com- 7 2 + 2/2 1 + 3/3 plexes encapsulated in hemagglutinating virus of Japan lipo- 14 3/3 1 + 2/2 1 + somes resulted in blue staining in the primate trabecular 30 2/3 ± 0/1 meshwork, although the staining was not as intense as in the 60 0/3 - rat trabecular meshwork (Figs. 2a, 2b). The injection of FITC-labeled phosphorothioate oligonucleotides encapsulated in hemagglutinating virus of Japan liposomes also re- after injection and persisted for up to 14 days (n = 3 at each sulted in high fluorescence in the primate trabecular mesh- time point, Table 1). The uninjected fellow eyes showed no work (Fig. 3a). However, there were fewer cells expressing /3-Gal activity in any anterior chamber tissue when examined /3-Gal than those showing the fluorescence. Neither the from 1 to 60 days after injection (data not shown). We also uninjected fellow eyes (w = 6) nor the eyes that were evaluated eyes injected with naked LacZ DNA, naked LacZ injected with naked LacZ DNA (ji — 6) showed any /3-Gal DNA-high-mobility group 1 complexes, LacZ DNA-high-mo- activity in any anterior chamber tissue, including the trabec- bility group 1 complexes encapsulated in liposomes without ular meshwork, when examined from 4 to 30 days after viral coats of hemagglutinating virus of Japan, or LacZ DNA injection (data not shown). LacZ expression in the primate without high-mobility group 1 encapsulated in hemagglutinat- trabecular meshwork lasted for as long as 2 weeks (n = 1 or ing virus of Japan liposomes. None of these groups showed any 2 at each time point; Table 2). There was no LacZ staining expression of LacZ in any anterior chamber tissue throughout in any other anterior chamber tissues, such as the cornea, the study period (n = 2 in each group at each time point; Fig. iris-ciliary body, or lens. A low-power dark field of eyes lc). transfected with 3 /xM FITC-labeled phosphorothioate oligo- Hemagglutinating virus of Japan liposomes seemed to fa- nucleotides showed that they were introduced into the cilitate the introduction of oligomers into the same types of trabecular meshwork but not into the iris-ciliary body (Fig. cells in rats, though high-mobility group 1 was not used (Figs. 3c). A high-power dark field of eyes transfected with 3 /xM Id, le, If)- Fluorescence was seen in the trabecular meshwork FITC-labeled phosphorothioate oligonucleotides showed and iris-ciliary body of eyes injected with 3 /xM FITC-labeled that the fluorescence was primarily in the corneoscleral and phosphorothioate oligonucleotides encapsulated in hemagglu- juxtacanalicular portions of the trabecular meshwork (Fig. tinating virus of Japan liposomes (Figs. Id, le) 1 hour after 3a). Autofluorescence of elastic fibers can be converted to a 23 injection, and it gradually decreased but was detectable 7 days dark red by counterstaining with erichrome black T. The after injection (n = 3 at each time, Table 2). In contrast, little fluorescence in the trabecular meshwork was unchanged by fluorescence was seen in eyes injected with the same concen- this treatment (data not shown), indicating that the fluores- tration of naked FITC-labeled phosphorothioate oligonucleo- cence was not derived from possible autofluorescence of 24 tides (n = 3 at each time, Fig. If). elastic fibers present in the trabecular meshwork. The There was no LacZ expression in rats injected with LacZ fluorescence in the trabecular meshwork seemed to be co- DNA-high-mobility group 1 complexes encapsulated in hem- localized with the nuclear staining by propidium iodide. agglutinating virus of Japan liposomes in nonocular tissues, This suggests that phosphorothioate oligonucleotides intro- such as the brain, liver, kidney, or aorta. Although some of duced by hemagglutinating virus of Japan liposomes were

TABLE 2. Time Course of Fluorescence Intensity in Rats and Monkeys Rats Monkeys

Naked HVJ*liposomes Naked HVJ liposomes

Time n\ 1% n I n I n /

1 hour 2/3 ± 3/3 3 + 1/2 ± 2/2 3 + 1 day 0/3 3/3 2 + 0/2 2/2 2 + 3 day 3/3 1 + 1/2 1 + 7 clay 2/3 •+• 0/2 14 day 0/3

*HVJ = hemagglutinating virus of Japan, in = number of eyes. intensity.

Downloaded from iovs.arvojournals.org on 09/23/2021 IOVS, March 1998, Vol. 39, No. 3 Introduction of DNA into Trabecular Meshwork 513

No animals died before the scheduled killing. The anterior chamber showed a blurred appearance just after the injection, AC but it became clear when observed the clay after. No animals showed signs of eye irritation such as redness, abnormal tear- ing, or too frequent blinking. No abnormal findings, except for surgical injury to the cornea, were observed with an operating microscope. There were no histologic signs of inflammation, such as infiltration of polymorphonuclear leukocytes or mono- nuclear cells, in any anterior chamber tissue. Apparent tissue damage, such as corneal endothelial swelling, endothelial de- tachment from the Descemet membrane, or abnormal vessel dilation in the iris-ciliary body, was not detected histologically.

DISCUSSION This study demonstrates that the introduction of a reporter gene and phosphorothioate oligonucleotides into the rat and primate trabecular meshwork is feasible if the hemagglutinating virus of Japan liposomes method is used. It has been shown previously that the viral envelope proteins mediate the fusion of liposomes with the cell membrane and that high-mobility group 1, a nonhistone nuclear protein, mediates the trafficking of the intracellularly introduced plasmid DNA into the nuclei.2526 These proteins also were necessary to achieve the introduction of transgenes into rat eyes because neither naked LacZ DNA, naked LacZ DNA-high-mobility group 1 complexes, LacZ DNA-high-mobility group 1 complexes encapsulated in liposomes without viral coats of hemagglutinating virus of Japan, nor LacZ DNA without high-mobility group 1 encap- AC sulated in hemagglutinating virus of Japan liposomes resulted in any expression of LacZ in any anterior chamber tissue (Fig. lc). The outcomes of gene transfer experiments may vary markedly from one species to another because of differences in anatomic structures and immune surveillance. Pre- vious experiments evaluating in vivo gene transfer into ocular tissues have used small animals, such as mice, rats, or rabbits.14"1627'30 However, for gene therapy, it would be SC more relevant to determine whether gene transfer into mammals closer to humans, such as nonhuman primates, is feasible. The present data showed gene transfer into nonhuman FIGURE 2. Introduction of LacZ DNA into the primate trabec- primate ocular tissues in vivo. The drainage system through ular meshwork (TM). LacZ expression in the trabecular mesh- the trabecular meshwork is particularly different between 1718 work 10 days after the injection of LacZ DNA-high-mobility higher primates and lower mammals. Lower mammals group 1 complexes encapsulated in liposomes with (a, b) or have reticulate and far less extensive trabecular meshwork without (c) hemagglutinating virus of Japan coats. AC, anterior and, except for rats and guinea pigs, have no Schlemm's chamber; SC, Schlemm's canal. Magnification: X125 (a, c); and canal.31'32 Therefore, the achievement of gene transfer into X280 (b). the rat trabecular meshwork shown in this study could provide a promising and inexpensive means for analyzing accumulated in the nuclei, the primary site of antisense the functions of the trabecular meshwork, but it does not action (Figs. 3a, 3b). In contrast, the injection of 3 /xM naked necessarily guarantee that cellular and molecular mecha- FITC-Iabeled phosphorothioate oligonucleotides resulted in nisms involved in glaucoma can be investigated or that little fluorescence in any anterior chamber tissue (Fig. 3d). somatic gene therapy of the trabecular meshwork to treat The fluorescence in eyes transfected by hemagglutinating glaucoma is feasible by using our method. Because the virus of Japan liposomes gradually decreased but was detect- architecture involved in the trabecular outflow pathway of able 3 days after injection (n — 1 or 2 in each group at each monkeys is closer to that of humans than any other species, time point, Table 2). the rhesus monkey seems to be one of the most relevant models for testing the efficacy and for evaluating characteris- Pathogenicity of Hemagglutinating Virus of Japan tics of vectors for gene transfer into the trabecular mesh- liposomes work. Throughout the study, all the animals appeared healthy and There seemed to be differences in the hemagglutinating showed no changes in their consumption of food and water. virus of Japan liposome-mediated gene transfer between rats

Downloaded from iovs.arvojournals.org on 09/23/2021 514 Hangai et al. IOVS, March 1998, Vol. 39, No. 3

FIGURE 3 Introduction of phosphorothioate oligonucleotides into the primate trabecuiar meshwork (TM). (a-c) Fluores- cence was seen in the trabecuiar meshwork of eyes transfected with fluorescein isothiocyanate (FITC)-labeled phosphorothioate oligonucleotides by hemagglutinating virus of Japan liposomes, primarily in the corneoscleral and juxtacanalicular sections (a), and it was colocalized with nuclear staining by propidium iodide (a, b). This fluorescence was unchanged by counterstaining with erichrome black T, indicating that the fluorescence was not derived from the autofluorescence of elastic fibers in the trabecuiar meshwork. Note the small amount of fluorescence outside the trabecuiar meshwork (c). (d) Little fluorescence was seen in the trabecuiar meshwork of eyes injected with naked FITC-labeled phosphorothioate oligonucleotides. SC, Schlemm's canal; I? iris; CB, ciliary body; and CM, ciliary muscle. Magnification: X60 (a, b, d); and X30 (c).

the anterior chamber, or in the number and distribution of proteins with sialic acids to which hemagglutinin-neuranirni- dase, an envelop protein of hemagglutinating virus of Japan, binds at the initiation of fusion.25'33'34 Another difference was that transfer of LacZ DNA into the primate trabecuiar meshwork seems to be less efficient than transfer into the rat trabecuiar meshwork, although phosphorothioate oligonucleotides were introduced effectively into the primate trabecuiar meshwork. There were fewer monkey trabecuiar meshwork cells transfected with LacZ DNA than those transfected with phosphorothioate oligonucleotides. In addition, they showed weaker blue staining than the rat trabecuiar meshwork cells. The reason is uncertain, but one possible explanation is that the chicken j3-actin promoter used in the present study may not work efficiently in the primate trabecuiar meshwork cells. It was reported that cultured primate trabecuiar meshwork cells were effectively transfected with reporter genes using adenoviral vectors under the control of the rous sarcoma virus promoter.16 Further efforts to increase the expression of a reporter gene, such as using more powerful promoters including those from a virus, will be necessary to use this system effectively in primates. The introduction of a reporter gene by intracameral injection of hemagglutinating virus of Japan liposomes was confined to the trabecuiar meshwork and the iris-ciliary body in rats and to the trabecuiar meshwork in rhesus monkeys at the concentrations used. In contrast, it was shown that the replication- deficient adenovirus was effective in transferring genes to a wider range of anterior chamber tissues, such as the trabecuiar meshwork, corneal endothelium, iris epithelium, and lens epithelium, when evaluated in mice and rabbits.1'1"16 The different results may be caused by the distinctions in the physical natures of adenoviral vectors and hemagglutinating virus of Japan liposomes that can lead to their different diffusions in the anterior chamber aqueous flow dynamics. Another possible explanation is that there may be large differences in the sus- ceptibility of anterior chamber tissues or cells to adenoviral vectors and hemagglutinating vims of Japan liposomes. For example, different distributions of receptors for adenovirus infection and proteins with sialic acids25'33'3'1 may cause the and primates. One difference was that LacZ DNA and phos- different spectra of gene deliveries into the anterior chamber phorothioate oligonucleotides were introduced into the iris- tissues. Furthermore, the differences in aqueous outflow dy- ciliary body of rats but not of rhesus monkeys. The reasons for namics between species may have influence on the times and this may be explained by the differences between the species concentrations it takes for vector particles to contact each in the trabecuiar architecture, in the aqueous flow dynamics in anterior chamber tissue.

Downloaded from iovs.arvojournals.org on 09/23/2021 IOVS, March 1998, Vol. 39, No. 3 Introduction of DNA into Trabecular Meshwork 515

The identification of glaucoma-related genes raises not 5. Garcia-Valenzuela E, Shareef S, Walsh J, Sharma SC. Programmed only the interest in the functions of the gene products but also cell death of retinal ganglion cells during experimental glaucoma. the possibility of controlling intraocular pressure by modulat- Exp Eye Res. 1995;6l:33-44. 6. Yue BY. The extracellular matrix and its modulation in the trabec- ing the gene expression. Gene transfer or the introduction of ular meshwork. Surv Opbthalmol. 1996;40:379-390. antisense phosphorothioate oligonucleotides into the primate 7. Tanihara H, Ohira A, Takahashi M, Honda Y, Suzuki S. Localization trabecular meshwork by hemagglutinating virus of Japan lipo- and possible gene expression of proteoglycan decorin in the tra- somes will provide an opportunity to study the glaucoma- becular meshwork. Curr Eye Res. 1995;l4:727-730. related genes. However, the transient expression of a reporter 8. Sheffield VC, Stone EM, Alward WL, et al. Genetic linkage of gene by hemagglutinating virus of Japan liposomes will limit familial open angle glaucoma to chromosome Iq21-q31. Nat their practical use as a vector for human ocular gene therapy Genet. 1993;4:47-50. because longer expression requires repeated injection, and this 9. Richards JE, Lichter PR, Boehnke M, et al. Mapping of a gene for autosomal dominant juvenile-onset open-angle glaucoma to chro- will increase the risk of complications such as infection. The mosome lq. Am]Hum Genet. 1994;54:62-70. development of plasmids that have a capacity for autonomous 10. Morissette J, Cote G, AnctiJ JL, et al. A common gene for juvenile replication or integration into the genome through homolo- and adult-onset primary open-angle glaucomas confined on chro- gous recombination or the exploration of artificial chromo- mosome lq. Am J Hum Genet. 1995;56:l431-l442. somes may increase the potential of this method.35"37 At this 11. Richards JE, Lichter PR, Herman S, et al. Probable exclusion of time, the transient expression of a reporter gene and the GLC1A as a candidate glaucoma gene in a family with middle-age- introduction of antisense oligonucleotides by hemagglutinating onset primary open-angle glaucoma. Ophthalmology. 1996; 103: 1035-1040. virus of Japan liposomes seem to be sufficient for investigating 12. Meyer A, Bechetoille A, Valtot F, et al. Age-dependent penetrance molecular mechanisms by which trabecular aqueous outflow and mapping of the locus for juvenile and early-onset open-angle resistance increases in open-angle glaucoma. Thus, hemagglu- glaucoma on chromosome lq (GLC1A) in a French family. Hum tinating virus of Japan liposomes could provide useful insights Genet. 1996;98:567-571. into the trabecular meshwork not only for the development of 13. Stone EM, Fingert JH, Alward WLM, et al. Identification of a gene somatic gene therapy but also for the exploration of novel that causes primary open angle glaucoma. Science. 1997;275:668- medical treatments of glaucoma. 670. 14. Abraham NG, da-Silva JL, Lavrovsky Y, et al. Adenovirus-mediated Targeting the pathologic changes within the trabecular heme oxygenase-1 gene transfer into rabbit ocular tissues. Invest meshwork is an attractive approach for glaucoma manage- Ophthalmol Vis Sci. 1995;36:2202-2210. ment. Antiglaucoma drugs currently used to lower intraocular 15. Budenz DL, Bennett J, Alonso L, Maguire A. In vivo gene transfer pressure do not target the pathologic changes within the tra- into murine corneal endothelial and trabecular meshwork cells. becular meshwork. Pharmacologic trials to change the subcel- Invest Opbthalmol Vis Sci. 1995;36:2211-2215- lular structure of the trabecular meshwork have been carried 16. Borrds T, Tamm ER, Zigler J Jr. Ocular adenoviiiis gene transfer varies in efficiency and inflammatory response. Invest Ophthalmol out using various agents such as cytochalasins, EDTA chelates, Vis Sci. 1996;37:1282-1293. hyaluronidases, and ethacrynic acids, although none of these 17. Rohen JW. The histologic structure of the chamber angle in pri- have yet been established as glaucoma therapy.38"41 These mates. Am] Opbthalmol. 196l;52:529. therapies that target the trabecular meshwork may lessen the 18. Rohen JW. The evolution of the primate eye in relation to the limitations of nonsurgical management of glaucoma. Somatic problem of glaucoma. In: Liitjen-Drecoll E, ed. Basic Aspects of gene therapy, which can deliver therapeutic proteins to the Glaucoma Research. Stuttgart: Schattauer Verlag; 1982:3-34. 19. Kato K, Nakanishi M, Kaneda Y, Uchida T, Okada Y. Expression of trabecular meshwork, would be a possible novel therapy for hepatitis B virus surface antigen in adult rat liver. Co-introduction altering the trabecular meshwork to treat glaucoma. of DNA and nuclear protein by a simplified liposome method. In summary, LacZ DNA and phosphorothioate oligonucle- JBiolCbem. 1991;266:336l-3364. otides were introduced into the rat trabecular meshwork and 20. Sihha ND, Striepeke S. Oligonucleotides with reporter groups iris-ciliary body and into the primate trabecular meshwork by attached to the 5'-terminus. In: Eckstein F, ed. Oligonucleotides the hemagglutinating virus of Japan-liposome method. Be- and Analogues: A Practical Approach. Oxford: 1RL Press; 1991: cause the transient and weak expression of a reporter gene in 202-205. 21. Kaneda Y. Virus (Sendai virus envelopes>mediated gene transfer. rhesus monkeys by the existing system limits its usefulness, In: CelisJE, ed. Cell Biology: A Laboratory Handbook. San Diego: further studies are needed to modify the liposomes and plas- Academic Press; 1994:50-57. mid vectors to achieve more efficient expression. Depending 22. Hangai M, Kaneda Y, Tanihara H, Honda Y. In vivo gene transfer on its improvement, this system has the potential not only to into the retina mediated by a novel liposome system. Invest Oph- open new avenues for glaucoma research but also to develop thalmol Vis Sci. 1996;37:2678-2685. nonviral somatic gene therapy of the trabecular meshwork to 23. Kittelberger R, Davis PF, Stehbens WE. An improved immunofluorescence technique for the histological examination of blood treat glaucoma. vessel tissue. Acta Histochem. 1989;86:137-l42. 24. Murphy CG, Yun AJ, Newsome DA, Alvarado JA. Localization of References extracellular proteins of the human trabecular meshwork by 1. Quigley HA. Number of people with glaucoma worldwide. Br J indirect immunofluorescence. Am] Ophthalmol. 1987;104:33- Opbthalmol. 1996;80:389-393. 43. 2. Shields MB. Textbook of Glaucoma. 3rd ed. Baltimore: Williams & 25. Okada Y. Factors in fusion of cells by HVJ. Curr Top Microbiol Wilkins; 1992:176-177. Immunol. 1969;48:102-128. 3. Bnibaker RF. Delayed functional loss in glaucoma. Am] Ophthal- 26. Kaneda Y, Iwai K, Uchida T. Increased expression of DNA coin- mol. 1996;121:473-483. troduced with nuclear protein in adult nit liver. Science. 1989;243: 4. Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, 375-378. Zack DJ. Retinal ganglion cell death in experimental glaucoma and 27. Bennett J, Wilson J, Sun D, Forbes B, Maguire A. Adenovirus after axotomy occurs by apoptosis. Invest Opbthalmol Vis Sci. vector-mediated in vivo gene transfer into adult murine retina. 1995:36:774-786. Invest Ophthalmol Vis Sci. 1994;35:2535-2542.

Downloaded from iovs.arvojournals.org on 09/23/2021 516 Hangai et al. IOVS, March 1998, Vol. 39, No. 3

28. Li T, Adamian M, Roof DJ, et al. In vivo transfer of a reporter gene 35. Masukata H, Satoh H, Obuse C, Okazaki T. Autonomous replication to the retina mediated by an adenoviral vector. Invest Ophthalmol of human chromosomal DNA fragments in human cells. Mol Biol VisSci. 1994;35:2543-2549. Cell. 1993:4:1121-132. 29. Mashhour B, Couton D, Perricaudet M, Briand P. In vivo adenovi- 36. Vitta-Pearlman VJ, Gunaratne PH, Chinault AC. Analysis of a repli- rus-mediated gene transfer into ocular tissues. Gene Ther. 1994; 1: cation initiation sequence from the adenosine deaminase region of 122-126. the mouse genome. Mol Cell Biol. 1993;13:5931-5942. 30. Masuda I, Matsuo T, Yasuda T, Matsuo N. Gene transfer with lipo- 37. Huxley C. Mammalian artificial chromosomes: A new tool for gene somes to the intraocular tissues by different routes of administration. therapy. Gene Ther. 1994;1:7-12. Invest Ophthalmol VisSci. 1996;37:19l4-1920. 38. Peterson WS, Jocson VL. Hyaluronidase effects on aqueous out- 31. Tripathi RC, Tripathi BJ. The mechanism of aqueous outflow in flow resistance. Quantitative and localizing studies in the rhesus lower mammals. Exp Eye Res. 1972;l4:73-79- monkey eye. Am J Ophthalmol. 197'4;77r:573-577'. 32. van-der-Zypen E. Experimental morphological study on structure and function of the nitration angle of the rat eye. Ophthalmo- 39- Kaufman PL, Barany EH. Cytochalasin B reversibly increases out- logica. 1977; 174:285-298. flow facility in the eye of the cynomolgus monkey. Invest Oph- 33- Homma M. Trypsin action on the growth of Sendai virus in tissue thalmol Vis Sci. 1977;l6:47-53. culture cells. I. Restoration of the infectivity for L cells by direct 40. Bill A. Effects of Na2 EDTA and alpha-chymotrypsin on aqueous action of tyrpsin on L cell-borne Sendai virus./ Virol. 1971;8:6l9- humor outflow conductance in monkey eyes, dps J Med Sci. 629. 1980;85:3H-319. 34. Scheid A, Caliguiri LA, Compans RW, Choppin PW. Isolation of 41. Melamed S, Kotas-Neumann R, Barak A, Epstein DL. The effect paramyxovirus glycoproteins. Association of both hemagglutinat- of intracamerally injected ethacrynic acid on intraocular pres- ing and neuraminidase activities with the larger SV5 glycoprotein. sure in patients with glaucoma. Am J Ophthalmol. 1992;113: Virology. l972;50:640-652. 508-512.

Downloaded from iovs.arvojournals.org on 09/23/2021