The authors have declared that no competing interests exist.
Conceived and designed the experiments: GSW MJ YHL YN NAR. Performed the experiments: GSW MJ YHL YN. Analyzed the data: GSW NAR. Contributed reagents/materials/analysis tools: GSW MJ NAR. Wrote the paper: GSW.
Unlike its constitutive isoforms, including neuronal and endothelial nitric oxide synthase, inducible nitric oxide synthase (iNOS) along with a series of cytokines are generated in inflammatory pathologic conditions in retinal photoreceptors. In this study, we constructed transgenic mice overexpressing iNOS in the retina to evaluate the effect of sustained, intense iNOS generation in the photoreceptor damage.
For construction of opsin/iNOS transgene in the CMVSport 6 expression vector, the 4.4 kb Acc65I/Xhol mouse rod opsin promoter was ligated upstream to a 4.1 kb fragment encoding the complete mouse cDNA of iNOS. From the four founders identified, two heterozygote lines and one homozygote line were established. The presence of iNOS in the retina was confirmed and the pathologic role of iNOS was assessed by detecting nitrotyrosine products and apoptosis. Commercial TUNEL kit was used to detect DNA strand breaks, a later step in a sequence of morphologic changes of apoptosis process.
The insertion and translation of iNOS gene were demonstrated by an intense single 130 kDa band in Western blot and specific immunolocalization at the photoreceptors of the retina. Cellular toxicity in the retinas of transgenic animals was detected by a post-translational modification product, tyrosine-nitrated protein, the most significant one of which was nitrated cytochrome c. Following the accumulation of nitrated mitochondrial proteins and cytochrome c release, marked apoptosis was detected in the photoreceptor cell nuclei of the retina.
We have generated a pathologic phenotype with sustained iNOS overexpression and, therefore, high output of nitric oxide. Under basal conditions, such overexpression of iNOS causes marked mitochondrial cytochrome c nitration and release and subsequent photoreceptor apoptosis in the retina. Therefore, the modulation of pathways leading to iNOS generation or its effective neutralization can be of significant therapeutic benefit in the oxidative stress-mediated retinal degeneration, a leading cause of blindness.
There are three known isoforms of nitric oxide synthase and all three isoforms generate nitric oxide (NO) by the catalytic conversion of arginine to citrulline
Consistent with the current trend of using the gene knockouts (KOs) to evaluate the function of target genes, iNOS KO has frequently been used in recent studies
Further, using cardiovascular systems, the respective functions of all three isoforms of NOS have been investigated in pharmacological studies with specific NOS inhibitors and also in studies with mice that lack iNOS isoforms. These studies concluded that there were always some elements of uncertainty, such as in pharmacological studies, the specificity of the NOS inhibitors continued to be an issue of debate, and while in each type of the NOS isoform-deficient mice, compensatory effects by other NOS isoform were frequently encountered
Intraocular injections of commonly used NO donors have been reported in rats; similar reports in mice, however, are scarce. An intravitreal injection of N-ethyl-2-(1-ethyl-2-hydroxy-2-nitrosohydrazino) ethanamine (NOC12) causes cell loss in GCL and thinning in IPL and INL, but no effect on ONL. NOC-12 is known to release a larger amount of NO spontaneously. Other studies using S-nitroso-N-acetylpenicillamine (SNAP) failed to demonstrate any retinal toxicity
As
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Association of Research in Vision and Ophthalmology. The protocol was approved by the Committee of the Ethics of Animal Experiments of the University of Southern California (Protocol Number: 11218). All surgery was performed under ketamine and xylazine anesthesia, and all efforts were made to minimize suffering of animals.
The opsin/iNOS transgene was constructed by ligating the 4.4 kb Acc65I/XhoI mouse rod opsin promoter upstream of the 4.1 kb fragment encoding the complete mouse cDNA of iNOS in pCMVSport6 expression vector (MGC mouse verified FL clone, mouse NOS2; Open Biosystems, Huntsville, AL). The mouse opsin promoter in pBluescript SKII was a kind gift from Dr. J. Chen (Cell and Neurobiology, University of Southern California)
The 4.4 kb Acc65I/Xhol mouse opsin promoter fragment was isolated and cloned into the Acc65I/Sall sites of pCMVSport6-iNOS. The polyadenylation signal was provided by the existing SV40 polyadenylation sequence which was immediately downstream of the iNOS cDNA. To release the transgene from the vector for microinjection, a double restriction enzymatic digestion employing Acc65I and Pvul was performed.
The identification of founders, breeding of funders to generate F1 lines, initial screening of homozygotes and heterozygotes, and test breeding of possible homozygotes to WT to establish the homozygosity were all performed according to the published methods
Approximately 0.5 cm long mouse tails were cut and placed in 200 µl tail lysis buffer containing 195 µl mouse tail direct lysis reagent (Viagen Biotechnology) and 4.8 µl proteinase k (Viagen Biotechnology) and incubated at 55°C for 26 hours. The lysed mixture was then heated to 85°C for 45 minutes to degrade the remaining proteinase k before PCR. TaqPCR Core Kit (Qiagen, Valencia, CA) was used to prepare the master mix, and 1–2 µl of supernatant from the tail lysis mixture was used for the PCR.
Primers for generating the 339 bp PCR fragment have been described in the first section of Materials and Methods.
The isolated retinas were combined in a group of four and subjected to protein isolation (including low-output sonication) in Tris buffer (250 µl) containing a mixture of protease inhibitors (Roche Diagnostics, Indianapolis, IN). No detergent was added. Only a 20 µg portion of total extracted cytoplasmic protein (400 µg/4 retinas) was used per lane in Western blot. The protein was then fractionated by SDS-PAGE and the electrophoresed proteins were transferred by a semi-dry transfer unit (Hoefer, Inc. San Francisco, CA) to polyvinylidine fluoride (PVDF) membrane (Millipore, Billerica, MA). Nonspecific sites were blocked with 0.1%Tween 20 in 50 mM Tris-buffered saline (pH 7.4) throughout the staining procedure. For iNOS immunoblots (7.5% gel), the primary antibody used was rabbit polyclonal anti-iNOS (BD Biosciences, San Jose, CA; catalogue #, 610332; lot #, 18483; and dilution, 1∶2000) and secondary antibody was goat anti-rabbit IgG conjugated with biotin (Vector Laboratories, Burlingame, CA; catalogue #, BA-1000; lot #, W-1002; and dilution: 1∶200). For probing β-actin, primary antibody was mouse monoclonal IgG (Santa Cruz Biotechnology, Santa Cruz, CA; catalogue #, sc-47778; lot #, C3012; and dilution, 1∶1000) and the secondary antibody was goat anti-mouse IgG conjugated with biotin (Vector Laboratories; catalogue #, BA9200; lot #, W0206; and dilution: 1∶200). For nitrotyrosine immunoblots (15% gel), the primary antibody was rabbit polyclonal anti-nitrotyrosine (Millipore; catalogue #, 06-284; lot #, DAM 1748585; and dilution, 1∶200) and the secondary antibody was the same as that used for iNOS. For cytochrome c blots (15% gel), the primary antibody was monoclonal anti-mouse cytochrome c (BD Biosciences; catalogue #, 556433; lot #, 00736; dilution: 1∶300) and the secondary antibody was goat anti-mouse IgG conjugated with biotin (Vector Laboratories; same as the secondary antibody used for β-actin. Species specificity for all of the antibodies used includes mouse. After enhancement with a complex of peroxidase-conjugated biotin and avidin (ABC kit, Vector Laboratories), visualization was carried out using chromogenic 3,3′-diamino-benzidine/NiCl2 reagent (Sigma, St Louis, MO). Visualization was often carried out by Pierce enhanced chemiluminescence (ECL) Western blotting substrate (Thermo Scientific, Rockford, IL) to compare the results with chromogenic detection. The specificity of polyclonal anti-iNOS from BD Biosciences (catalogue #, 610332) was routinely checked for its specificity using a commercially available mouse macrophage + IFNγ/LPS lysate (BD Biosciences, catalogue # 611473; concentration, 1 mg/1 ml). The iNOS induced by these activated mouse macrophages has been cloned and characterized in the past
Enucleated eyes devoid of corneas were fixed in 4% paraformaldehyde (PFA) at 4°C and then cryo preserved in 30% sucrose overnight (4°C) before embedding in optimal cutting temperature compound (OCT). The enucleated, intact eyes were also embedded in OCT first and the frozen sections were fixed with 4% PFA for 10 minutes at room temperature. Throughout the study, 10 µm sections were used for the staining. For localizing iNOS, non-antigenic sites were blocked with 2% goat serum and 1% BSA for 20 minutes. Slides were then incubated with the primary antibody, rabbit polyclonal anti-iNOS (BD Biosciences, the same primary used for the Western analysis; dilution, 1∶100) overnight at 4°C, and then with the secondary antibody, Alexa Fluor 568 goat anti-rabbit IgG (Molecular Probes/Invitrogen, Eugene, OR; catalogue #, A11036; lot #, 757102; and dilution, 1∶500) 1 hour at room temperature. Zeiss 510 laser scanning confocal microscope was used for the visualization. For localizing nitroyrosine, the primary antibody used was rabbit polyclonal anti-nitrotyrosine (Millipore; the same antibody as that for the Western probing; dilution, 1∶100) and secondary antibody was Alexa Fluor 488 goat anti-rabbit IgG (Molecular Probes/Invitrogen; catalogue #, A111034; lot #, 760000; and dilution, 1∶1000). All Alexa Fluor secondary antibodies were incubated 1 hour at room temperature. Zeiss 510 laser scanning confocal microscope was used for the visualization. For immunolocalization of nitrotyrosine, two control procedures were used to ascertain the specificity of the primary antibody: (a) primary antibody was replaced by PBS; (b) primary antibody was blocked by reacting with 5 mM commercial nitrotyrosine (Sigma) before staining.
For light microscopy, eyes were enucleated without any injury and placed in Davidson's fixative (95% ethyl alcohol, 33.3 ml; 10% buffered formalin, 22.2 ml; glacial acetic acid,11.1 ml; and distilled water, 33.3 ml) for 24 hours before subjecting to the routine embedding. The steps of tissue processing are: 1) dehydration using graded strength of ethanol; 2) use of xylene as a clearing agent; 3) infiltration with paraffin and 4) orienting the tissue sample in paraffin and allowing it to solidify. To cut paraffin blocks, a thermostatically controlled water bath is used to float out tissue ribbons after sectioning. A drying oven (at 60 degree) was used to melt the paraffin before H&E staining.
Serial paraffin sections were cut through most of the eye, and selected sections were stained with hematoxylin and eosin (H&E). For each phenotype, multiple sections from at least two eyes collected from different animals were examined.
The opsin/iNOS transgene (
No developmental abnormality was noted in either homozygote or heterozygote offsprings. However, the breeding efficiency of homozygote to homozygote was noted to be less than normal. For all the experiments in the characterization of iNOS insertion and subsequent detection of retinal degeneration, 5 to 6 months old iNOS transgenic mice were used. The reason for selecting predominantly 5- to 6-month-old animals for experiments are twofold: 1) From the results reported by the existing literature in similar studies, P150 was reported to be the age when effects of the mutation become apparent in mouse opsin promoter-directed expression of simian virus 40 tumor antigen gene
To confirm the presence of iNOS in the retina, the Immunoblots were carried out on the total retinal homogenates probed with iNOS polyclonal antibody. A single, intense band was detected at 130 kDa (iNOS) using β-actin as loading control (
Following SDS-PAGE, the proteins were transferred to PVDF membrane for detecting with antibodies. Upper panel was probed with rabbit polyclonal anti-iNOS as primary antibody and goat anti-rabbit IgG conjugated with biotin as secondary antibody. For detecting β-actin in the lower panel, the primary antibody was mouse monoclonal anti-β-actin and goat anti-mouse Ig G conjugated with biotin. Lane 1: heterozygote, line a; lane 2: homozygote; lane 3: heterozygote, line b; and lane 4: C57BL/6 control mice.
Translated iNOS protein level was further evaluated by localization of iNOS in the retina. In the confocal visualization, the opsin promoter-driven iNOS expression was intensely and specifically localized in photoreceptor inner segment (IS) and outer plexiform layer (OPL) specifically (
Rabbit polyclonal anti-iNOS (primary antibody) and Alexa Fluor 568 goat anti-rabbit IgG (secondary antibody) were used for the staining. A: retina from transgenic homozygote; and B: retina from C57BL/6 control. Retinal layers marked are: GCL: ganglion cell layer; INL: inner nuclear layer; OPL: outer plexiform layer; ONL: outer nuclear layer; and IS: photoreceptor inner segments. Note the intense staining of iNOS specifically in the IS and OPL and low intensity, dispersed staining in INL and IPL. In the control, only some low grade, non-specific staining is seen.
Following establishment of the iNOS insertion, the pathogenic effect of iNOS was determined by evaluating NO/peroxynitrite-driven nitrosative stress in the iNOS transgenic retina. We sought specifically for the tyrosine-nitration in proteins, a stable oxidative/nitrosative post-translational nitration of tyrosine residues in the retina. In nitrotyrosine immunoblots of iNOS homozygotes and heterozygotes, a major nitrated band was seen near 15 kDa, indicative of nitrated cytochrome c
The total proteins were electrophoresed on 15% polyacrylamide gel and were probed with rabbit polyclonal anti-nitrotyrosine and biotinylated goat anti-rabbit IgG antibodies. Following enhancement with ABC kit, chromogenic visualization was used for the detection. Enhanced chemiluminescence (ECL)-based visualization was also carried out routinely for the comparison. Western blot analyses were carried out in triplicate and representative results are shown. Lane 1: heterozygote line a; lane 2: heterozygote line b; lane 3: C57BL/6 control; and lane 4: homozygote. Although there are several low-intensity tyrosine-nitrated bands in the background, the major nitrated protein pattern appears to be similar in all zygotes, with a major band near 16 kDa (A: nitrated cytochrome c monomer) and a doublet near 30 kDa (B: nitrated cytochrome c dimer). Further confirmation of these bands is presented in
The identification of the15 kDa band as nitrated cytochrome c was further shown by the two Western blots from the same membrane, but was probed separately with two different antibodies: lanes 1, 2, and 3, with monoclonal cytochrome c antibody and lanes 4, 5, and with polyclonal anti-nitrotyrosine detecting only a small part of the protein structure containing nitrated tyrosine (
The retina proteins were extracted with Tris buffer containing only protease inhibitors and no detergent. Total retina proteins were separated with15% polyacrylamide gel before transferring to PVDF membrane. Lanes 1, 2, and 3 were immunoblotted with monoclonal mouse cytochrome c antibody; lane 1: heterozygote line a; lane 2: homozygote; and lane 3: heterozygote line b. Lanes 4, 5, and 6 were immunoblotted with nitrotyrosine antibody; lane 4: homozygote; lane 5: heterozygote line a, and lane 6: heterozygote line b. A: cytochrome c; B: nitrated cytochrome c; C: cytochrome c dimer and D: nitrated cytochrome c dimer. Nitrated cytochrome c was presumably displaced from its binding site at the electron transport chain and was then released to the cytosol after a gentle mechanical disruption of the mitochondrial outer membrane by sonication.
The immunolocalization of nitrated proteins indicated that the prominent nitrated protein sites to be IS, and OPL (
In TUNEL staining, numerous apoptotic cells stained with red TMR were found exclusively in the nuclei of photoreceptor outer nuclear layer (ONL) and were absent in other layers (
Antibodies used for the confocal immunolocalization are rabbit polyclonal anti-nitrotyrosine and Alexa Fluor 488-conjugated goat anti-rabbit IgG (green). Propidium iodide (red) was used for the nuclear staining. The intense localization was seen specifically and uniformly in the photoreceptor inner segments (IS) and outer plexiform layer (OPL). These locations are known mitochondria-rich areas. A: retina section from homozygote; B: retina section from control. Retina layers labeled are: INL: inner nuclear layer; OPL: outer plexiform layer; ONL: outer nuclear layer; IS: photoreceptor inner segments and OS: photoreceptor outer segments.
Retinal apoptosis was detected using
After confirming the apoptosis in ONL, the morphologic changes in 6-month old homozygotes were evaluated. Using Davidson's fixative and undisturbed whole eyes, the H & E staining revealed no apparent major photoreceptor disruption. The retina was still generally preserved, except for some minor disorganized retinal layers, uneven thickness of ONL, and occasional retinal detachment (data not shown). We are continuing to evaluate the kinetics of photoreceptor dystrophy at different time periods on hetero- and homozygotes older than 6 months.
In this study, we used the mouse photoreceptor rod opsin promoter to direct overexpression of mouse iNOS in the photoreceptors. The efficacy of this promoter has previously been shown to be efficient and capable of inserting higher levels of transgene in the photoreceptors
The iNOS-derived pathologic effects were evaluated by detecting the oxidative/nitrosative products, focusing on the post-translationally modified tyrosine- nitrated proteins. In confocal localization, the major sites of protein nitration were in photoreceptors IS and OPL where the mitochondria are densely populated
For evaluating the formation of protein tyrosine-nitration, protein extraction was performed using lysis buffer (containing Tris and protease inhibitors only), suitable for extracting only the soluble cytoplasmic proteins, exclusive of cytoskeletal- or membrane-bound proteins. No detergent was added in this lysis buffer. From extracted cytosolic proteins, intense nitrated protein bands were detected at 15 kDa and 30 kDa, consistent with cytochrome c monomer and dimer respectively. The identity of these bands was further confirmed by blotting the same cytosolic proteins with 1) cytochrome c antibody detecting both cytochrome c monomer and dimer, and 2) nitrotyrosine antibody detecting the nitrated tyrosine structure in any protein. The results indicated that the 15 and 30 kDa bands were indeed cytochrome c and that it was also tyrosine-nitrated (
Cytochrome c is a multi-functional enzyme involving in both life and death of the cell. It participates in electron transfer reaction as part of the mitochondrial electron transport chain and is thus an important molecule for energy production process
A functional cytochrome c normally binds electrostatically to both mitochondrial respiratory complex III and IV in the mitochondrial inner membranes. It is therefore, stable to gentle mechanical disruption of the retina, but sensitive to detergents that are added into the lysis buffer and capable of digesting cellular membranes
It is now generally accepted that in the mitochondrial death pathway, a key step in the apoptotic cascade involves the release of cytochrome c into the cytosol where it binds with apoptotic protease-activating factor 1 (Apaf-1). Binding of cytochrome c results in an increased affinity of the complex for dATP leading to the formation of apoptosome
The retinal photoreceptor apoptosis occurs in various retinal diseases
Although low constitutive levels of previously known nNOS and eNOS are expressed in iNOS transgenic mice, there is evidence to suggest that these isoforms do not confer a substantial effect in the retinal degeneration. The reasons are summarized as follows: 1) Constitutive nNOS/eNOS and iNOS have distinctively different functions. Neuronal NOS has been implicated in synaptic plasticity, central control of blood pressure, and neurotransmission; whereas, eNOS-derived NO is a physiological vasodilator
In spite of these prominent degenerative effects of iNOS displayed, there is not yet a gross alteration of retinal morphology observed in the photoreceptors of iNOS transgenic mice. In a future study, visual functional tests, such as ERG, should be carried out on these transgenic animals. One likely reason for not seeing the photoreceptor loss could be the upregulation of protective mechanisms, such as crystallins
In summary, we have generated a pathologic phenotype with sustained iNOS expression in the photoreceptors. These transgenic mice allow us to evaluate the genuine iNOS effect totally discerned from the immune mediators by which iNOS is normally induced and associated in inflammation. At this stage of investigation, our results revealed that cytochrome c was tyrosine-nitrated and released from its normal binding site in the mitochondrial inner membrane. In the past, the occurrence of protein nitration through the iNOS oxidative/nitrosative pathway has been well documented