`
`l99l. 1009—10l4
`
`CYAN EXHIBIT 1036
`
`Current
`Eye
`Research
`
`Chemotactic activity of the peroxidized retinal membrane lipids in experimental
`autoimmune uveitis
`
`
`Hiroshi Goto, Guey—Shuang Wu, David C.Gritz, Lily R.Atalla and Narsing A.Rao
`
`
`Doheny Eye Institute and the Department of Ophthalmology, University of Southern California School of
`Medicine, Los Angeles, CA, USA
`
`ABSTRACT
`We investigated the mechanism for amplification ()f
`intraocular inflammation in rats with experimental
`autoimmune uveitis by examining the chemotaxis
`potentials of peroxidized lipids extracted from the
`retinas. Utilizing thin layer chromatography, we found
`that the peroxidized products isolated from the inflamed
`retinas were fatty acid hydroperoxides that corresponded
`to the autooxidized products from commercial methyl
`docosahexaenoate, with Rf values ranging from 0.30 to
`0.37. These were not demonstrated in similar
`preparations from normal retinas or in unoxidized
`docosahexaenoate. Boyden chamber assay revealed that
`the hydroperoxides isolated from inflamed eyes and the
`products of oxidized methyl docosahexaenoate possessed
`significantly higher chemotactic activity than did the
`retinal lipids isolated from normal eyes (P < 0.01).
`These findings may help to explain the mechanism of
`inflammatory amplification induced by peroxidized
`retinal lipids that is seen in this animal model of uveitis.
`
`INTRODUCTION
`
`The inflammatory response is composed of events that
`
`can be grouped according to type: vascular, cellular, or
`
`humoral. The hallmark of the cellular component of
`
`acute inflammation is the neutrophil. Migration and
`
`accumulation of neutrophils is governed largely by a
`
`arachidonic acid metabolites, are released by the
`
`inflammatory cells in an attempt to destroy the target of
`
`the inflammation. Recently, oxygen metabolites, such as
`
`superoxide, hydrogen peroxide, hypochlorous acid and
`
`hydroxyl radicals, have been shown to cause tissue
`
`damage via peroxidation of the membrane lipids, and are
`
`responsible also for the amplification of a variety of
`
`inflammatory conditions, including those involving the
`
`uvea(1,2).
`
`We have investigated the mechanism of the
`
`perpetuation and amplification of inflammation by
`
`analyzing the chemotactic activity of the peroxidized
`
`lipids extracted from retinal tissue. We hypothesize that
`
`these lipids, following the peroxidation that accompanies
`
`inflammation, contribute to the perpetuation of
`
`chemotaxis and to the amplification of inflammation that
`
`characterizes experimental autoimmune uveitis.
`
`MATERIALS AND METHODS
`
`Indu ti n f x
`
`rimental
`
`toimmune v i is EA
`
`group of mediators, called chemotactic agents, that are
`
`Animal experimentation was performed in accordance
`
`liberated at the site of the inflammatory initiation.
`
`with the Declaration of Helsinki and the Guiding
`
`Known chemotactic agents, such as arachidonic acid
`
`Principles in the Care and Use of Animals (DHEW
`
`metabolites, have been implicated in inflammation
`
`Publication, NIH 80-23).
`
`throughout the body, including uveitis. Temporally,
`
`Following anesthesia with intramuscular ketamine
`
`inflammation is thought to consist of an initiation phase,
`
`hydrochloride and xylazine hydrochloride, Lewis rats
`
`followed by a perpetuation or amplification phase that is
`
`(weighing approximately 175 g each) were given a single
`
`believed to be associated with the inflammatory
`
`hind foot-pad injection of 50 pg bovine S—antigen in
`
`mediators. However, these mediators may not be the
`
`complete Freund’s adjuvant containing 0.2 mg heat killed
`
`only factors involved in amplification of the
`
`Mycobacterium. Animals were observed for 15 days, at
`
`inflammatory response.
`
`which time they were killed with an overdose of
`
`During the inflammatory reaction, a variety of
`
`intramuscular ketamine hydrochloride and xylazine
`
`substances, including oxygen metabolites, proteases, and
`
`hydrochloride.
`
`
`
`Received on May 22. 1991: accepted on September 5, 1991
`
`
`C) ()xford [Iniversny Press
`
`1009
`
`H‘Ifl H T13
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcarecombyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`
`
`Current
`Eye
`Research
`
`
`et al (6). This procedure has been shown to give nearly
`
`Following enucleation, retinas of rats were dissected
`
`quantitative conversion of phospholipids and triglycerides
`
`from the globes and the membrane lipids were extracted
`
`to fatty acid methyl esters (6,7). In a typical experiment,
`
`using the method of Folch et al(3). Briefly, the combined
`
`to a solution of crude lipids (8—10 mg obtained from 12
`
`retina tissues from 6 eyes were homogenized in 6 ml of
`
`chloroform/ methanol (2:1) containing 0.5 mg of
`
`eyes) dissolved in 0.2 ml of methanol, 0.3 ml of METH-
`PREP II was added. The reaction mixture was then
`
`butylated hydroxytoluene per 100 ml of solvent. The
`
`stirred under nitrogen at room temperature for 30
`
`pooled extracts were then washed with 1.2 ml of water
`
`and centrifuged. The solvent was then evaporated from
`
`minutes. At the end of this period, 0.8 ml of water was
`added and the mixture was then extracted twice with 2 ml
`
`the organic layer and total lipids was obtained as residue.
`
`each of ethyl acetate. The organic layer was separated by
`
`In vitro preparation of peroxidized retinal lipids was also
`
`centrifugation, pooled and evaporated under nitrogen to
`
`performed by isolating the lipids from normal control rats
`
`obtain usually 4—5 mg of total faty acid methyl esters.
`
`and incubating these lipids with a radical-generating
`
`TLC was carried out using precoated silica gel 60, Merck
`
`system consisting of 200 uM of 2,2’-azobis (2-
`
`plates (0.25 mm thick) and a solvent system consisting of
`
`amidinopropane) hydrochloride (AAPH) (Polysciences,
`
`petroleum ether/diethyl ether/acetic acid (70:30:1). The
`
`Inc., Warrington, PA) dissolved in Hanks’ balanced salt
`
`spots on the plates were visualized by dipping in a
`
`solution (HBSS) (4).
`
`solution of 3% cupric acetate in 8.5% phosphoric acid
`
`Commercial methyl docosahexaenoate (Nu-Chek-
`
`and then charring at 140°C. The extent of oxidation of
`
`Prep, Elysian, MN) was used in its pure and oxidized
`
`the commercially available methyl docosahexaenoate was
`
`forms. The oxidized methyl docosahexaenoate was
`
`also monitored by TLC using the same solvent system.
`
`prepared by air oxidizing the commercial pure methyl
`
`9W
`
`docosahexaenoate (Nu Chek Prep, Inc., Elysian, MN) at
`
`Human polymorphonuclear leukocytes (PMNs) were
`
`room temperature for 5 to 7 days. The extent of
`
`isolated from heparinized peripheral blood utilizing
`
`autoxidation was monitored by measuring the diene
`
`dextran sedimentation and centrifugation of Ficoll-
`
`conjugation formed in the sample. For this
`
`Hypaque gradients, as previously described(8).
`
`measurement, an aliquot of the sample was dissolved in
`
`Chemotactic responses were measured in a modified
`
`1 ml of ethanol and the absorption range of 200-400 nm
`
`multiwell Boyden chamber(8). A portion of lipids to be
`
`was recorded using a Shimadzu spectrophotometer
`
`model UV-160. For estimating the quantity of
`
`conjugated dienes formed in the sample, absorbance at
`233 nm was measured and molar extinction coefficient of
`
`used for Chemotactic activity measurement was initially
`dissolved in 95% ethanol and the solvent was then
`
`removed via evaporation under nitrogen. The residue
`
`was suspended in 200 pl of HBSS, and this solution was
`
`25,200 was used for the calculation (5). In a typical
`
`placed in the lower compartment of the chamber. A
`
`experiment, 50 mg of methyl docosahexaenoate was
`
`suspension of PMNs (2x106/m1) in HBSS with 0.5%
`
`oxidized and 40 to 50% of the oxidized lipid could be
`
`bovine serum albumin was placed in the upper
`
`obtained in 7 days.
`
`compartment. Two chamber wells were divided by a
`
`To confirm the presence of fatty acid
`
`single nitrocellulose filter with a 5.0 pm pore diameter.
`
`hydroperoxides and hydroperoxide-derived hydroxy fatty
`
`acids in the inflamed retinal tissue, thin layer
`
`chromatography (TLC) was performed. Total lipids
`extracted from the normal and inflamed retinas were
`
`The chamber was incubated at 37° in 5% CO2 for 60
`minutes. Filters were removed, fixed in alcohol, and
`
`stained with hematoxylin and eosin. The Chemotactic
`
`activity was quantitated as the average number of cells
`
`transesterified to obtain fatty acid methyl esters.
`
`per high power field (x400) migrating completely through
`
`Transesterification was performed by reacting the lipids
`
`the filter in four different locations(8). Cell counts were
`
`with METH-PREP II (0.2 M m-trifluoromethylphenyl-
`
`performed by two observers, masked as to the filter
`
`trimethylammonium hydroxide in methanol, Alltech
`
`Associates, Deerfield, IL) using the method of van Kuijk
`
`group. Each observer independently examined all of the
`filters and chose four different locations on each that
`
`
`
`1010
`
`RIGHTS
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcarecombyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`
`
`Current
`Eye
`Research
`
`
`they felt were representative of the entire filter. All
`
`During reduction by glutathione peroxidase in tissue and
`
`assays were carried out in triplicate.
`
`RESULTS
`
`Detection of lipid peroxidation
`
`The peroxidized products isolated from inflamed retinas
`
`were compared with the autooxidized products obtained
`
`from the commercial methyl docosahexaenoate by
`
`transesterification by METH-PREP II in processing,
`
`most of the hydroperoxy fatty acid methyl esters are
`
`converted to hydroxy fatty acid methyl esters. These
`
`hydroperoxide spots were not recognizable in fatty acid
`
`preparations from the normal retinas or in unoxidized
`
`methyl docosahexaenoate.
`h m
`i
`ivi
`
`measuring conjugated dienes and by TLC. The major
`
`Boyden chamber assay revealed that the peroxidized
`
`products formed in both cases were fatty acid
`
`retinal lipids possessed much greater chemotactic ability
`
`hydroperoxides, which were seen as multiple spots with
`
`than did retinal lipids from normal rats (Table 1). The
`
`Rf values ranging from 0.30 to 0.37 (Figs. 1 and 2).
`
`
`
`lipids isolated from inflamed eyes (10 mg/ml) were more
`
`chemotactic than were those isolated from normal eyes
`
`(P < 0.01). The same extent of increased chemotaxis was
`
`observed in the samples of lipids that were oxidized in
`
`vitro using the AAPH radical generating system. The
`
`
`5
`
`1‘.
`
`1
`
`I
`
`‘::1
`
`Figure 1. Thinla er chromatogram of retinal lipids
`isolated from both, normal and EAU rats. Channel 1,
`oxidized soybean phosphatidylcholine standard; channel
`2, total fatty acid methyl esters isolated from the retina
`and choroid of control animals; channel 3, total fatty
`acid methyl esters isolated from the retina and choroid
`of EAU animals; A, unchanged fatty acid methyl esters
`(Rf=0.57-0.65; B, hydroperoxy fatty acid methyl ester
`(Rf: 0.30-0.;37 C, bydroxy fatty acid methyl esters
`(Rf: 0. 19-0.27; D, cholesterol. Note the presence of
`hydroperoxides and bydroxy fatty acids1n channel 3 and
`an absence of both hydropléyroxides and hydroxyfatty
`acid in channel 2.
`
`
`
`B
`
`1
`
`2
`
`Fig. 2. Thin layer chromatogram of air--oxidized methyl
`docosahexaenoate Channel 1, commerciall pure 22: 6;
`channel 2, air--oxidized 22.6; A,Bunchanged att acid
`methyl esters ERf= 0.58-0.67;;13,hydroperoxy atty acid
`methyl esters Rf=0.29-0.38 .
`
`
`
`1011
`
`H'I'GHTB
`
`1’
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcare.combyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcare.combyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`Current
`Eye
`Research
`
`
`assays using reduced concentrations of the same lipids
`
`chemotactic factors are produced that serve to attract
`
`(1 mg/ml) yielded similar increases in chemotaxis. Using
`
`PMNs and other inflammatory cells to the site of
`
`the commercially available docosahexaenoic acid, the
`
`inflammation. In the case of EAU induced by S—antigen,
`
`chemotactic activity of the air-oxidized fatty acid was
`
`the infiltration of PMNs and macrophages can be seen in
`
`found to be 3-fold higher than that of the unoxidized
`
`the retina, as well as in the uvea. These leukocytes then
`
`form (P < 0.01).
`
`DISCUSSION
`
`Inflammatory processes, including uveitis, appear to
`
`release inflammatory mediators, including oxygen
`
`metabolites(9).
`
`In inflammation, several biologically active lipids,
`such as the arachidonic acid metabolites, have been
`
`occur in two steps:
`
`induction, followed by perpetuation
`
`implicated as chemotactic factors. In this study, we have
`
`shown that the lipids isolated from the retinas of animals
`or amplification. In the amplification phase of EAU,
`
`
`Table 1. Chemotactic ability as measured by Boyden chamber analysis
`
`Source of chemoattractant
`
`Concentration
`
`Chemotaxis
`
`p value?
`
`
`
`of lipids (mg/ml) (Net PMNs/HPF‘)
`
`Retinal lipid extracted
`
`from normal rats (control)
`
`Retinal lipid extracted
`
`from EAU rats
`
`Retinal lipid extracted
`
`from normal rats and oxidized
`
`via AAPH in Vitro
`
`AAPH alone
`
`Unoxidized, pure
`
`methyl docosahexaenoate§
`
`10.0
`
`1.0
`
`10.0
`
`1.0
`
`10.0
`
`1.0
`
`0323.4
`
`0.3 :01
`
`28.2:0.4
`
`3.7: 1.3
`
`<0.01
`
`<0.01
`
`28.7:6.6
`
`2.0: 1.0
`
`.0:3.6
`
`<0.01
`
`< 0.01
`
`(control for oxidized form)
`
`1.04.1 : 1.9
`
`Oxidized
`
`methyl docosahexaenoate§
`
`0.5
`
`1.0
`
`0.5
`
`2.3 11.2
`
`13.4: 1.2
`
`7.2: 1.2
`
`<0.05
`
`<0.05
`
`‘PMNs/HPF = Number of polymorphonuclear neutrophils per high power field. Shown as mean
`1- standard deviation.
`TCompared with control.
`§Refers to commercially available methyl docosahexaenoate.
`
`1012
`
`RIGHTS
`
`"f
`
`
`
`Current
`Eye
`Research
`
`
`with EAU have chemotactic ability. These chemotactic
`
`contribution from other chemotactic factors, in particular
`
`properties could be due to the peroxidized retinal lipids,
`
`oxygenated arachidonic products.
`
`since lipids isolated from control animals also showed
`
`chemotactic abilities when they were exposed in vitro to
`
`ACKNOWLEDGMENTS
`
`an effective radical generating system, AAPH. This
`
`Supported in part by grant EY 05662 from the National
`
`system is a well known radical initiator, and the
`
`spontaneous decomposition of this azo compound to free
`
`Institutes of Health and by an unrestricted grant from
`Research to Prevent Blindness, Inc., New York, NY.
`
`radicals is followed by reactions with oxygen molecules,
`
`Ann Dawson, medical editor, reviewed this manuscript.
`
`rapidly yielding peroxyl radicals(10). The oxidized
`
`polyunsaturated fatty acids have been found to possess
`
`CORRESPONDING AUTHOR
`
`chemotactic properties( 1 1).
`
`In the outer segment of photoreceptors, the
`
`Narsing Rao, Doheny Eye Institute, University of
`Southern California, 1355 San Pablo Street,
`
`membrane phospholipids contain more than 50 mole
`
`Los Angeles, CA 90033.
`
`percent of docosahexaenoic acid (22:6), and only a small
`
`percentage of arachidonic acid( 12). Arachidonic acid is
`
`derived from dietary omega-6 series and is the precursor
`
`of prostaglandins and thromboxanes. Docosahexaenoic
`
`acid, on the other hand, is derived from dietary omega-3
`
`series, and cannot take part in the synthesis of
`
`arachidonic acid pathway products. In EAU, the reactive
`
`oxygen radicals released by PMNs are capable of
`
`oxidizing retinal polyunsaturated fatty acids and it has
`
`been shown the 22:6 is oxidized during this process(9).
`
`The peroxidized 22:6 was found to be mostly
`
`hydroperoxides(9, 13). As one of the detoxification
`
`mechanism, tissue glutathione converts hydroperoxides to
`
`the corresponding hydroxy fatty acids. However, this
`
`protective mechanism might have been destroyed by the
`
`initial influx of reactive oxygen species(9). In the
`
`peroxidation of 22:6, there are ten possible isomers of
`
`hydroperoxides, resulting from the five sets of 1, 4-dienes
`
`in the molecule. Using mass spectrometry/gas
`
`chromatography, we have positively identified five major
`
`isomers of hydroperoxide derived
`
`hydroxydocosahexaenoic acids from the inflamed
`
`retinas(9).
`
`Lipid peroxidation products are known to lead to
`
`cell edema and increased vascular permeability(14). We
`
`have found that the most abundant polyunsaturated fatty
`
`acid in photoreceptor membranes, 22:6, upon subjecting
`
`to PMN-mediated peroxidation could be chemotactic.
`
`This, undoubtedly, contributes to further elicitation of
`
`PMNs and thus amplifies the inflammatory process. In
`
`REFERENCES
`1.
`Ward, P.A., Till, G.O., Hatherill, J.R., Annesley,
`TM. and Kunkel, R.G. (1985) Systemic
`complement activation, lung injury, and products
`of lipid peroxidation. J. Clin. Invest. E, 517-527.
`Rao, N.A., Sevanian, A., Fernandez, M.A.S.,
`Romero, J.L., Faure, J.P., de Kozak, Y., Till, GO.
`and Marak, G.E. Jr. (1987) Role of oxygen
`radicals in ex erimental allergic uveitis. Invest.
`Ophthalmol.
`is. Sci. 3, 886-892.
`Folch, J., Lees, M. and Soloane Stanley, G.H.
`(1957) A sim 1e method for the isolation and
`urification 0 total lipides from animal tissues. J.
`iol. Chem. 2_2_6, 497-509.
`Yamamoto, Y., Niki, E., Kami a, Y., Shimasaki,
`H. (1984) Oxidation of lipids.
`. Oxidation of
`phosphatidylcholines in homogeneous solution
`and in water dispersion. Biochim. Biophys. Acta
`125, 332-340.
`Bue e, J.A. and Aust, SD. (1978) Microsomol
`lipi peroxidation. Methods Enzymol. 2, 302-
`310.
`Van Kuijk, F.J.G.M., Thomas, D.W., Stephens,
`R.J. and Dratz, EA. (1985) Gas chromatography-
`mass spectrometry method for determination of
`hospholipid peroxides: Transesterification to
`orm methyl esters. Free Radic. Biol. Med. 1, 215-
`225.
`
`McCreary, D.K., Kossa, W.C., Ramachandran, S.
`and Kurtz, RR. (1978) A novel and rapid method
`for the preparation of methyl esters for gas
`chromatography: Ap lication to the
`determination of the atty acids of edible fats and
`oils. J. Chromatogr. Sci. 111, 329-331.
`Cates, K.L., Ray, CE. and Quie, P.G. (1978)
`Modified Boyden chamber method of measuring
`polymorphonuclear leukocyte chemotaxis. In
`"Leukocyte Chemotaxis" (eds. Gallin, J.I. and
`Quie, P.G.). Pp. 67-71. Raven Press, New York,
`NY.
`
`Rao, NA. (1990) Role of oxygen free radicals in
`retinal damage associated with ex erimental
`uveitis. Trans. Am. Ophthalmol. oc. 88, 797-851.
`Terao, K. and Niki, E. (1986) Damage to
`biological tissues induced by radical initiator 2,2’-
`azobis (2-amidinopropane) dihydrochloride and
`conducted to refine the system, thus to evaluate the
`
`
`this study, the contribution from other chemotactic
`
`membrane lipids was not excluded, the studies are being
`
`10.
`
`1013
`
`RIGHTS
`
`"f
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcare.combyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`
`
`Current
`Eye
`Research
`
`
`11.
`
`12.
`
`13.
`
`14.
`
`its inhibition by chain-breaking antioxidants. J.
`Free Radic Biol Med. 2, 193-201.
`Turner, SR. Campbell, JA. and Lynn, W. S.
`(1975) Polymorphonuclear leukocyte chemotaxis
`toward oxidized lipid components of cell
`membranes. J. Exp. Med. 151, 1437-1441.
`Stone, W..,L Farnsworth, CC. and Dratz, EA
`(1979) A reinvesti ation of the fatty acid content
`of bovine, rat and reg retinal rod outer segments.
`Exp. Eye Res. 2_8, 387g—397.
`Porter,N.A. (1984) Chemistry of lipid
`peroxidation. Methods Enzymol. 105, 273—282.
`Southorn, RA. and Powis, G. (1988) Free radicals
`in medicine. 1. Chemical nature and biologic
`reactions. Mayo Clin. Proc. fl, 381-389.
`
`1014
`
`H‘I‘GHTB
`
`'1;-
`
`personaluseonly.
`
`
`
`CurrEyeResDownloadedfrominformahealthcare.combyGarryHandelmanon06/05/13For
`
`
`
`
`
`
`
`