Distribution of membrane phospholipids in the rabbit neural retina, optic nerve head and optic nerve

hr. /. Biochenr.

Pergamon

0020-711X(94)00061-1

Cell Bid. Vol. 27, No. I, pp. 21 -28. 1995 Copyright :(; 1995 Elsevier Science Lid Printed in Great Britain. All rights reserved

13S7-2725/95

$9.50 t 0 00

of Membrane P al Retina, Optic N Optic Nerve* CYNTI-IIA A. M. GREINER,’ JACK V. GREINER,‘“? CHARLES D. LEAHY,’ DAVID B. AUERBACH,’ MIRIAM D. MARCUS,’ LAURA H. DAVIES,’ WILMA RODRIGUEZ,’ THOMAS GLONEK3 ‘Schepens Eye Research Institute, ‘Department of Ophthalmology, Harvard Medical School, 20 Staniford St, Boston, Massachusetts and ‘Magnetic Resonance Laboratory, Midwestern University, Chicago, Illinois, U.S.A. Since diseases of the neural retina and optic nerve can rest& in alteration of biologkal membranes, this study determines similar&s and diRerences in the membrane phospholipid content of the neural retina, optic nerve head, and optic nerve to serve as baseline data. Neural rethm, optk nerve bead, and optic nerve were dissected, isdated as 5 sets from 20 rabbits and frozen in liquid N,. Separate pooled-tissue extracts were prepared for each set of tissues and pho#~m+31 nuclear magnetic resonance (31PNMR) analyses performed. Ten pbosphcdip~lds were quantified (respectiveneural retina, optic nerve head, and optic nerve moie % are given for the 5 major pbosphoiipids detected): phosphatidykholiue (PC), 44.61,27.67, 26.00, PC plaamalogen or alhylacyl PC (CPLIP); phosphatidyKnositol (PI); spbingomyeiin (SMh pboaphatidylserine (I’S), 12.63,14.77,15.W, ~~~y~~~i~ (PE), 21.21,9.59, 8.69; PE plasma&en (EPLAS), 11.07,30.%, 33.93; an un&deutifiad(euhuown) H(u) at the chemiial-sbift vahte of 0.13 ppm; diphosphatidylglyceroerol (DPG); aud phosphatidk acid (PA), 0.46,2.92,1.57. Sign&ant differences between the various tissueswere determined by the one-way analysis of variance, using a Scheffe range value of P < 0.05. The neural retina in all phospboiipids detected except for the uncharacterized (unhnown) phosph&pid was signilkantly digerent from the optic nerve head tissue. The optic nerve head was slgnKkautly diRerent from the optic nerve in PC, CPLIP, PE, EPLAS, U, DPG, and PA. The data provide a baselinefor studies on pathologically changed neural retina, optic nerve head, and optic nerve. Keywords: Retina Optic nerve head Optic nerve Phospboiipids “P NMR

about the changes that occur in ocular diseases (e.g. glaucoma). Clinically, changes in glaucoma are evidenced by structural changes in the optic nerve head, monitored by ophthalmoscopy, and functional changes, monitored by examination of visual fields. These clinical changes are an indicator of biological membrane alteration that may be due to chronic relative elevation in intraocular pressure or changes secondary to alteration in blood circulation. Retinal phospholipids (Broekhuyse, 1968; Dreyfus et al., 1971; Horrocks, 1972; Ansell, 1973; Johnston and Hudson, 1974; Grafstein et al., 1975; Urban et al., 1975; Dreyfus et al., 1975; Dorman et al., 1977; Currie et al., 1978; Alberghina et al., 1982) and phospholipids of

INTRODUCTION

Changes in the structure and function of biological membranes can be monitored by changes in the tissue lipid molecular profile, particularly the phospholipid profile. Because the phospholipids comprise the cellular membranes that propagate neuronal signals, changes in phospholipids may be important in learning more *This study was presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Sarasota, FL, May 6, 1993. Supported in part by a grant from the Cytosol Foundation, Cytosol Laboratories, Inc., Braintree, MA. ~To whom all correspondence should be addressed. (Received 7 June 1994; accepted 17 October 1994). 21

22

Cynthia A. M. Greiner YI al

optic nerve tissue (Currie et al., 1978) have been reported. (The phospholipids of the optic nerve head have never been reported.) These earlier studies were performed on lipid fractional spots obtained by thin-layer chromatographic techniques. With the establishment of a method for analysis of tissue extracts utilizing highresolution 3’P nuclear magnetic resonance (NMR) (Meneses and Glonek, 1988), two important advancements have occurred: (1) increased numbers of phospholipids can be determined owing to the greater resolving power of NMR, and (2) precise quantitation is possible exhibiting a high degree-of-confidence level (Meneses and Glonek, 1988; Meneses et al., 1989; Merchant and Glonek, 1990; Meneses et al., 1993). The purpose of the present study was to (1) determine the phospholipid profiles of the neural retina, optic nerve head, and optic nerve, and (2) determine similarities and differences in the optic nerve head tissue that can be differentiated from the optic nerve proper or the neural retina. METHODS

Surgical

Eyes from New Zealand rabbits (n = 20) weighing between 5-6 kg were enucleated following a lethal dose of sodium pentabarbital (n = 40 eyes). Following enucleation the extraocular tissues were removed from the optic nerve and globe. Neural (sensory) retina, optic nerve head, and optic nerve tissues were dissected using an operating microscope. The distal 6 mm of the optic nerve was excised by cutting tangential to the sclera. The optic nerve head was comprised of that portion of the optic nerve that extended from the scleral wall to the retina, such that, upon removal, a full thickness hole remained in the globe. The globe was then opened anteriorly and the neural retina harvested carefully so as to avoid vitreous and ciliary body tissues. Phospholipid extraction procedure

Five samples consisting of 8 frozen tissue specimens each were pulverized to a fine powder with a stainless steel mortar and pestle chilled with liquid nitrogen. A simple modified Folch extraction (Folch et al., 1957) of retinal, optic nerve head, and optic nerve phospholipids was performed in which the backwashing step utilized potassium (ethylenedinitrilo)-tetraacetic acid (EDTA), 0.2 M in EDTA, pH 6.0 (Meneses

et al., 1993). The homogeneous tissue powder (average, 0.6 g/sample) was then added to 20 ml of chloroform-methanol (2/l, v/v). The homogenate, exhibiting only one liquid phase, was filtered through Whatman No. 1 filterpaper into a separatory funnel. The extract was washed with a 4 ml volume of 0.2 M KEDTA pH 6.0 and allowed to separate thoroughly for 24 hr in a separatory funnel. The chloroform phase was recovered and evaporated using a rotary evaporator at 37°C. The analytical medium for “P NMR phospholipid analysis was a hydrated chloroform-methanol NMR reagent (Meneses and Glonek, 1988; Glonek, 1994) specifically designed for the quantitative determination of phospholipids by 3’P NMR (Meneses and Glonek, 1988; Meneses et al., 1989; Edzes et al., 1992; Meneses et al., 1993; Glonek 1994). “P Nuclear magnetic resonance spectroscopy

A multinuclear GE 500NB spectrometer system operating at 202.4 MHz for 3’P was employed. Tissue extract samples were analyzed in standard 10 mm NMR sample tubes during 21 hr periods. Analysis was performed using proton broad-band decoupling to eliminate ‘H-3’P NMR multiplets (Meneses and Glonek, 1988). Chemical-shift data in units of ppm are reported relative to the usual standard of 85% inorganic orthophosphoric acid (Mark et al., 1967; Tebby and Glonek, 1991). The primary internal standard was the naturallyoccurring phosphatidylcholine (PC) resonance at -0.84 ppm (Meneses and Glonek, 1988). Identification of resonance signals was performed according to methods previously described (Greiner et al., 1981; Barany and Glonek, 1982; Glonek, 1994). Spectrometer analytical scan parameters were as follows: pulse sequence, one pulse; pulse width, 18 psec (45” spin-flip angle); acquisition delay, 500 psec; number of acquisitions, 50,000; free-induction decay size, 4 K channels; acquisition time, 2.05 set; sweep width, 1000 Hz. An exponential multiplication time-constant introducing 0.6 Hz computed line broadening was applied to enhance spectral signal-to-noise ratios. Resonance signal areas, chemical-shift measurements, and spectral curve analysis (used to resolve signals that overlapped) were calculated using software of the spectrometer’s computer system. Phospholipid concentrations in relative phosphorus molar percentages were computed from the phospholipid resonance

Phospholipids in retina and optic nerve

signal areas from each of the tissue specimen extracts. To compensate for relative saturation effects among various phosphorus signals detected in a single “P NMR spectral profile, the NMR spectrum must be standardized against measured amounts of tissue-profile metabolites wherever these are known. The procedures for carrying out this calibration, so that an accurate quantitative measurement is obtained from the 3’P spectral profile, have been described (Barany and Glonek, 1982; Meneses and Glonek, 1988; Merchant and Glonek, 1990; Edzes et al., 1992; Meneses et al., 1993). In addition, spectral curve resolution was employed to compute the relative signal areas of overlapping resonances in order to improve the integration of closely positioned signals (Sachedina et al., 1991). Phospholipids determined and metabolic indexes calculated

The following phospholipids were detected: phosphatidic acid (PA), diphosphatidylglycerol (DPG), an uncharacterized phospholipid (U), ethanolamine plasmalogen (EPLAS), phosphatidylethanolamine (PE), phosphatidylserine (PS), sphingomyelin (SM), phosphatidylinositol (PI), choline plasmalogen or alkylacylphosphatidylcholine (CPLIP), PC. In addition to measuring differences in tissue phospholipid composition, the analyses include metabolic indexes computed from groups of phosphoiipids, ratios of phospholipids and ratios of groups of phospholipids. These indexes compare and contrast phospholipids and provide more pathway-specific metabolic interrelations. The Indexes are mathematical devices that are useful for interpreting metabolic profile data. They are not an absolute measure, and further, they may not be supported by known metabolic processes. For example, the outside : inside ratios were inferred from erythrocytes and viruses. These have not been measured directly in ocular tissues, and it is known from other tissues that some fraction of phosphatidylcholine and sphingomyelin may exist on both sides of plasma membranes. The use of these indexes does not imply that all PC, for example, resides on the outer membrane leaflet. The computed metabolic indexes for the phospholipids are defined as follows: OUTSIDE, PC-t SM; INSIDE, PE+ PS; LEAFLET, OUTSIDE/INSIDE (Rothman and Lenard, 1977); SM/PC; SM/PE; SM/EPLAS; SM/PS; SM/PA; PC/PE; PC/EPLAS; PC/PS (Rietveld

23

et al., 1986); PC/PA (Smaal et al., 1986); PE/PS; PE/PA; EPLAS/PS; EPLAS/PA; PLASA, CPLIP + EPLAS; PLASB, PC + PE; UNSAT, PLASA/PLASB; UNSATC, CPLIP/PC; UNSATE, EPLAS/PE; LECITHIN, PC + CPLIP; PE + EPLAS; LECITHIN/ CEPHALIN, CEPHALIN; CHOLA (all choline-containing phospholipids), PC + CPLIP + SM; CHOLB (all other phospholipids), PI + PS + PE + EPLAS + DPG + PA; CHOLINE, CHOLA/CHOLB; ANIONIC/NEUTRAL (ratio of anionic phospholipids to neutral-ionic phospholipids), (PA + DPG + PS + PI)/(EPLAS + PE + SM -ICPLIP -t PC); PARATIO, PA/(PE + PS + PI + PC). Rationales for the use of these indexes have been discussed previously (Merchant et al., 1990; Merchant et al., 199 la, b, c; Sachedina et al., 1991; Merchant et al., 1993; Liang et ul., 1993). Data reductions and statistical analysis

Phospholipid concentrations were determined through integration of the phospholipid resonance signals detected from each of the tissue specimen extracts. The relative mole fraction of each signal contributing to a given spectral profile was then calculated as a percentage of the total spectral integral. Subsequent to this calculation of the relative phospholipid concentration values, the phospholipid metabolic indexes were calculated for each tissue specimen. From these computed relative phospholipid concentrations, mean relative mole percentages were calculated (Nie et al., 1975) for all resonances that were detected in all of the specimens of each tissue fraction: neural retina, optic nerve head, optic nerve. Similarly, tissue means were computed for all of the metabolic indexes constructed from the phospholipid relative mole percentages. Initially, the three tissue groups were compared at the level of the individual phospholipid or index values by an analysis of variance. For those mean phospholipid or index values where significance was determined to exist (F probability, < 0.05), post hoc simple contrasts were applied. Simple contrasts employed the Scheffe comparison procedure, with a P --c0.05 accepted as significant. Under most conditions, analysis of variance requires the assumption that the underlying variances between tested means are equal. For those resonances where significance was found to exist, homogeneity of variance was confirmed using Cochran’s C and the Bartlett-Box F tests (Nie et al., 1975).

24

Cynthia A. M. Greiner P[ ui. RESULTS

The -“P NMR spectral profile of the neural retina, optic nerve head, and optic nerve (Fig. 1) shows the presence of four major (EPLAS, PC, PS, SM) and several minor phospholipids (PA, DPG, PE, PI), including two uncharacterized phospholipids (U, CPLIP). The mean relative mole percentages of each phospholipid in the neural retina, optic nerve head, and optic nerve profiles are presented (Table 1). The choline and ethanolamine phospholipids comprise 80.4% of the neural retina phospholipid compliment, 78.6% of the optic

nerve head, and 80.2% of the optic nerve. The only other phospholipid that contributes substantially to the phospholipid profile of all three tissues is phosphatidylserine. From the 10 identified phospholipids, 29 computed indexes were calculated (Table 2). These theoretical parameters, given as ratios of individual or grouped phospholipids, were generated to compare phospholipids or groups of phospholipids, and, thereby, provide more pathway-specific metabolic interrelations for discussion. DISCUSSION

Considering both phospholipid levels (Table 1) and phospholipid metabolic indexes (Table 2), the neural retina, optic nerve head, and the optic nerve are significantly different Retina from each other. Quantitatively and qualitatively, the phospholipid composition of the optic nerve head is more similar to the optic nerve and considerably different from the neural retina. There are particular differences that differentiate the optic nerve head from the optic nerve JL proper. The PA level is elevated in the optic PC Optic nerve head nerve head relative to both the neural retina and the optic nerve. The detection of substantial PA EPLAS is a probable indicator of a phospholipase D activity in these three tissues. The detected PA is not an artifact of the extraction, since the lyso ,s M phospholipid derivatives are not detectable, precluding chemical breakdown during the CPLIP PI extraction. PARATIO is an index of the phospholipase L D equilibrium associated with the removal of the polar-head-group ester to generate PA. [It is Optic nerve not, however, an index of the phospholipase D activity responsible for the polar head group exchange that is known to occur in plants (Dawson, 1967; Yang et al., 1967), e.g. PI/PC]. The index is elevated in the optic nerve, indicating the presence of a phospholipase D activity in this neural tissue. Similarly, there is evidence v4 for the presence of a phospholipase D activity in I, 1 PPM another neural tissue, the sciatic nerve, with 0 -0.5 -1.0 myelin being the major locus of this activity (Eggen and Eichberg, 1992). Such an activity Fig. I. “P NMR nuclear magnetic resonance phospholipid profile of the neural retina, optic nerve head, and optic is characteristic of tissues that are growing nerve: PA, phosphatidic acid; DPG, diphosphatidylglycerol; and where the membrane phospholipid compoU, uncharacterized phospholipid; EPLAS, ethanolamine sition is in flux. The elevated PA level may be plasmalogen; PE, phosphatidylethanolamine; PS, phosphaan indication of increased metabolic activity, tidylserine; SM, sphingomyelin; PI, phosphatidylinositol; perhaps associated with a change in the axoCPLIP, choline plasmalogen or alkylacylphosphatidylcholine; PC, phosphatidylcholine. plasmic flow.

25

Phospholipids in retina and optic nerve Table I, “P nuclear magentic resonance phospholipid profiles of the neural retina, optic nerve head, and optic nerve Mole % detected phospholipid (Means f SD) Phosholipid PA DPG U EPLAS PE PS SM PI CPLIP PC

Chemical shift (wm) 0.36 0.18 0.13 0.11 0.06 -0.05 -0.09 -0.33 -0.78 -0.84

Neural retina 0.46 + 0.04 1.67 +O.lO 1.10+0.09 I I .07 + 0.26 21.21 + 0.36 12.63 + 0.40 2.88 + 0.12 3.30 f. 0.46 1.07+0.16 44.61 + 0.54

Optic nerve head 2.92 + 0.74 f 1.01 * 30.96 + 9.59 f 14.77 + 9.55 * 1.95 + 0.82 + 27.67 +

0.19* 0.23* 0.17 0.48* 0.07* 0.29* 0.14* 0.24* 0.08* 0.22*

Optic nerve 1.57 +O.l6t$ ---tt 1.54 +O.l6t$ 33.93 f 0.52jq 8.69 + 0.25Q 15.09 + 0.54t 9.80 i 0.24t 1.63 &O.l3f 1.36 + O.l7t$ 26.40 + 0.45tS

Abbreviations: PA, phosphatidic acid; DPG, diphosphatidylglycerol (cardiolipin); U, uncharacterized phospholipid; EPLAS, ethanolamine plasmalogen; PE, phosphatidylethanolamine; P’S, phosphatidylserine; SM, sphingomyelin; PI, phosphatidylinositol; CPLIP, choline plasmalogen or alkylacylphosphatidylcholine; PC, phosphatidylcholine. *Optic nerve head compared to neural retina (P < 0.05). tOptic nerve compared to neural retina (P < 0.05). SOptic nerve compared to optic nerve head (P < 0.05).

Two other notable differences are the presence of detectable amounts of DPG in the optic nerve head but not in the optic nerve proper and a level of an unknown phospholipid (CPLIP) that is reduced relative to both the optic nerve and the neural retina. DPG is an anionic phospholipid often associated with calcium ion translocation (Philipson et al., 1980). The metaindex bolic that applies is ANIONIC/NEUTRAL, which is significantly elevated in the optic nerve head relative to both the optic nerve and the retina. No significant difference in the ANIONIC/NEUTRAL index exists between the neural retina and the optic nerve, however. Other significant differences in phospholipid levels are reduced amounts of the unknown at 0.13 ppm and EPLAS and elevated levels of PE and PC in the optic nerve head relative to those of the optic nerve. When compared to the neural retina, both the optic nerve head and the optic nerve are reduced in PE, while exhibiting a compensatory enhancement in ethanolamine plasmalogen [indexes PLAS A (CPLIP+EPLAS), PLAS B (PC+PE), UNSAT (PLAS AjPLAS B), UNSAT C (CPLIP/PC), UNSAT E (EPLAS/ PE), LECITHIN, CEPHALIN, LECITHIN/ CEPHALIN, CHOLA, CHOLB, CHOLINE]. The enhancement of the plasmalogen over its diacyl analogue may reflect a more reducing environment in the nerve relative to that of the

neural retina, in which formation of the more chemically reduced plasmalogen enol-ether is favored. The ratios of the inner membrane leaflet phospholipids (PS, PE, EPLAS, and, perhaps, phospholipid U (Edzes et al., 1992) if the initial interpretation of this unknown proves correct) to the outer leaflet membrane phospholipids (SM, PC, and, perhaps, CPLIP) define membrane asymmetry (Rothman and Lenard, 1977). The computed metabolic indexes relevant to this feature are: OUTSIDE, INSIDE, LEAFLET, SM/PC, SM/PE, SM/EPLAS, SM/PS, PC/PE, PC/EPLAS, PC/P& PE/PS, EPLAS/PS. The indexes SM/PA, PC/PA, PE/PA, and EPLAS/ PA also apply, although it is not known on which lea&t of the membrane bilayer PA ordinarily resides. While numerous significances exist among the three tissues examined, the quantitative variances are small between the optic nerve head and the optic nerve and much larger between the optic nerve head and the neural retina. This may be anticipated from the close relationship between the optic nerve head and the optic nerve. The notable exceptions are all of those indexes defining systems involving PA. The retina is enriched in PC and PE and diminished in SM and EPLAS relative to both the optic nerve head and the optic nerve. These compensatory differences, while statistically

26

Cynthia A. M. Greiner et al. Table 2. 3’P Nuclear magnetic resonance phospholipid indexes of neural retina, optic nerve head, and optic nerve Index value (Means & SD) Index OUTSIDE INSIDE LEAFLET SMjPC SM/PE SM/EPLAS SMJPS SM/PA PC/PE PC/EPLAS PC/PS PC/PA PE/PS PE/PA EPLAS/PS EPLAS/PA PLASA PLASB UNSAT UNSATC UNSATE LECITHIN CEPHALIN LECITHIN/CEPHALIN CHOLA CHOLB CHOLINE ANIONIC/NEUTRAL PARATIO

Neural retina

Optic nerve head

47.49 + 0.48 33.84 + 0.39 1.40 + 0.020 0.065 f 0.003 0.136 + 0.007 0.260 k 0.012 0.228 + 0.008 6.31 + 0.56 2.10 f 0.051 4.03 f 0.132 3.53 + 0.101 7.94 If: 8.98 1.68 + 0.070 46.49 & 3.11 0.88 & 0.041 24.26 k 1.45 12.14 f 0.22 65.83 i 0.53 0.185 & 0.004 0.024 f 0.004 0.52 & 0.005 45.68 + 0.57 32.29 + 0.60 I .42 k 0.041 48.56 + 0.51 50.34 _+ 0.43 0.965 + 0.018 0.223 k 0.006 0.0056 + 0.0005

37.23 + 0.36* 24.37 + 0.30 I .53 + 0.028* 0.345 + 0.003* 0.996 _+0.012* 0.309 4 0.008* 0.647 + 0.021* 3.28 + 0.24* 2.89 + 0.020 0.89 + 0.017* 1.87 + 0.045* 9.50 + 0.69* 0.65 & 0.014* 3.29 f 0.23* 2.10 + 0.037* 10.64 f 0.84* 31.78 & 0.45* 37.27 f 0.27* 0.853 + 0.015* 0.030 * 0.003 3.23 + 0.065* 28.49 ) 0.23* 40.56 + 0.45* 0.70 + 0.011* 38.05 f 0.37* 60.94 f 0.37* 0.624 + 0.010* 0.259 + 0.007* 0.0542 + 0.0037*

Optic nerve 36.21 + 0.61?$ 23.77 2 0.75t 1.52 + 0.071t 0.371 + O.O08t$ 1.130 + 0.059tj 0.289 k 0.003tS 0.651 + 0.04lt 6.3 I _+0.64$ 3.04 jl 0.134tIf 0.78 k 0.01 It 1.75 f 0.080t 17.00 _+ 1.8lt 0.58 + 0.014Q 5.58 + 0.40t 2.25 _+0.117tf 21.84 k 2.18tS 35.29 f 0.5OQ 35.09 k 0.28t$ 1.006 f 0.017tS 0.052 k 0.007tS 3.91 *o.l7lt$ 27.76 + 0.32tS 42.62 & 0.30tS 0.65 & 0.007tj 37.57 & 0.5l.t 60.90 & 0.42t 0.617 + 0.013t 0.228 4 O.OlO$ 0.0302 + 0.003 ItI

Abbreviations: SM, sphingomyelin; PC, phosphatidylcholine; PE, phosphatidylethanolamine; EPLAS, ethanolamine plasmalogen; PS, phosphatidylserine; PA, phosphatidic acid; CPLIP, choline plasmalogen or alkylacylphosphatidylcholine; PI, phosphatidylinositol; DPG, diphosphatidylglycerol. Index definitions: OUTSIDE, PC+SM; INSIDE, PE+PS; LEAFLET, OUTSIDE/ INSIDE; PLASA, CPLIP+ EPLAS; PLASB, PC+ PE; UNSAT, PLASAIPLASB; UNSATC, CPLIP/PC; UNSATE, EPLAS/PE; LECITHIN, PC +CPLIP; CEPHALIN, PE + EPLAS; CHOLA (all choline-containing phospholipids), PC +CPLIP+ SM; CHOLB (ah other phospholipids), PI+PS+PE+EPLAS+DPG; CHOLINE, CHOLA/CHOLB; ANIONIC/NEUTRAL, (DPG+ PS + PI)/(EPLAS+ PE+ SM + CPLIP + PC); PARATIO, PA/(PE + PS + PI + PC). *Optic nerve head compared to neural retina (P c 0.05). TOptic nerve compared to neural retina (P < 0.05). SOptic nerve compared to optic nerve head (P < 0.05).

significant and large, are quantitatively small when compared through the LEAFLET index, indicating a similar balance in the polarity of the membranes among all three tissues. The PI concentrations in the optic nerve head and the optic nerve are relatively low for a mammlian tissue (Meneses and Glonek, 1988; Edzes et al., 1992; Liang et al., 1992; Liang et al., 1993), including other ocular tissues (Sachedina et al., 1991), but are similar to low PI levels detected in lenses from a variety of species (Merchant et al., 1991 b; Meneses

et al., 1990; Meneses et al., 1989; Feldman and Feldman, 1965; Broekhuyse, 1968; Glonek et a/., 1990). A feature of this 3rP NMR method is its ability to detect previously undetected and unknown phospholipids, such as phospholipids CPLIP and U (Table 1). The precise identity of the uncharacterized phospholipids is not known. CPLIP has been identified both as the choline plasmalogen (Meneses and Glonek, 1988), which contains the enol-ether functional group at the glycerol l-carbon position, and as

2-l

Phospholipids in retina and optic nerve

the corresponding alkylacylphosphatidylcholine (Edzes et al., 1992), which can be thought of as a derivative of the choline plasmalogen that has been further reduced to an alkyl function at the glycerol 1-carbon position. These assignments are not certain because of the inability, thus far, of producing chemically-pure grade samples of both compounds that are not crosscontaminated. A corresponding analogue of the platelet activating factor (alkenylacetylPC), the alkylacetylPC, is known to be an active biochemical (Lee and Snyder, 1985); however, neither the alkylacylPC or the alkylacetylPC are proven constituents of neural retina, optic nerve head or optic nerve tissues. It is suspected (Edzes et al., 1992), that the phospholipid labeled U is the corresponding alkylacylphosphatidylethanolamine, but this assignment is even less certain than those for the corresponding choline phospholipids. It should be noted, however, that the assignment of resonance U to the alkylacylphosphatidylethanolamine is consistent with data reported for the chemical and spectroscopic properties of an unknown phospholipid at 0.13 ppm in eye tissues (Glonek et al., 1990; Meneses et al., 1990; Merchant et al., 1990; Sachedina et al., 1991; Merchant et al., 1991b), as well as other human tissues (Merchant et al., 1991a, c; Merchant et al., 1993; Seijo et al., 1994). While the exact chemical identity of U and CPLIP are not known, these newly recognized phospholipids have characteristic and unique chemical shifts in the ,‘P NMR spectrum that permit their secure spectral recognition and precise quantitation in neural retina, optic nerve head, and optic nerve tissue spectroscopic phospholipid profiles. The use of indexes (Table 2) in this analysis provides a mechanism for interpreting metabolic profile data. The indexes represent sums and ratios of individual or grouped phospholipids that describe chemical functional groupings. For example, LECITHIN and CEPHALIN represent biochemical properties, LEAFLET and ANIONIC/NEUTRAL biochemical processes or pathways, EPLASjPE and similar indexes provide insight into known and unrecognized cellular compositions and processes. The fact that an index cannot be interpreted in terms of a particular known process, e.g., SM/PC, does not detract from its value as an index, since the index may represent a biochemical function unique to the tissue

under observation and important for eventual comparisons of neural retinal and optic nerve pathology as well as comparisons with other tissues. REFERENCES Alberghina M., Karlsson J. 0. and Giuffrida A. M. (1982) Rapid migration of inositol phospholipids with axonally transported substances in the rabbit optic pathway. J. Neurochem.

39, 223-221.

Ansell G. B. (1973) Phospholipids and the nervous system. In Form and Function of Phospholipids. (Edited by Ansell. Hawthorne, Dawson), 2nd edn, pp. 377-422. Elsevier. Amsterdam. Barany M. and Glonek T. (1982) Phosphorus-31 nuclear magnetic resonance of contractile systems. In Methods in Enzymology. (Edited by Frederiksen and Cunningham), Vol. 85B, pp. 624-676. Academic Press Inc., New York. Broekhuyse R. M. (1968) Phospholipids in tissues of the eye. I. Isolation, characterization and quantitative analysis by two-dimensional thin-layer chromatography of diacyl and vinyl-ether phospholipids. Biochim. Biophys. Acta 152, 307-31s. Currie J. R., Grafstein B., Whitnall M. H. and Alpert R. (1978) Axonal transport of lipid in goldfish optic axons. Neurochem.

Res. 3, 479-492.

Dorman R. V., Dreyfus H., Freysz L. and Horrocks 1. A. (1977) Ether lipid content of phosphoglycerides from the retina and brain of chicken and calf. Biochim. Biophys. Acta

486, 55-59.

Dreyfus H., Urban P. F., Edel-Harth S. and Mandel P. (1975) Developmental patterns of gangliosides and of phospholipids in chick retina and brain. J. Neurochem. 25, 245-250. Dreyfus H., Urban P. F. and Mandel P. (1971) Distribution des phosphatides et marquage du phosphore tipidique et du phosphate mineral de retines de rat et de veau. Biochimie

53, 567-569.

Edzes H. T., Teerlink T., van der Knaap M. S. and Valk J. (1992) Analysis of phospholipids in brain tissue by “P NMR at different compositions of the solvent system chloroform-methanol-water. Magn. Reson. Med. 26, 46-59. Eggen B. J. L. and Eichberg J. (1992) Phorbol estermediated stimulation of phosphohpase D activity in sciatic nerve from normal and diabetic rats. J. Neurochem. 59, 1467-1473. Feldman G. L. and Feldman L. S. (1965) New concepts of human lenticular lipids and their possible role in cataracts. Invest. Ophthalmol. Vis. Sri. 4, 1622166. Folch J., Lees M. and Sloane Stanley G. H. (1957) A simple method for the isolation and purification of total hpids from animal tissues. J. Biol. Chem. 226, 497--509. Glonek T., Greiner J. V. and Lass J. H. (1990) Application of “P NMR to eye research. In Principks and Applications to Biomedical Research. (Edited by Pettegrew N. M R.). pp. 1577203. Springer-Verlag, New York. Glonek T. (1994) P-31 NMR in the analysis of extracted tissue phospholipids. In Utilization of P-31 NMR Spectral Properties Analy,sis

in Compound

Characterization

and Structural

(Edited by Quin and Verkade). VCH Publishers. New York.

28

Cynthia A. M. Greiner er u/.

Grafstein B., Miller J. A., Ledeen R. W., Haley J. and Specht S. C. (1975) Axonal transport of phospholipid in goldfish optic system. Exp. Neural. 46, 261-281. Greiner J. V., Kopp S. J., Sanders D. R. and Glonek T. (198 I) Organophosphates of the crystalline lens: a nuclear magnetic resonance spectroscopic study. Inoesf. Ophthalmol. Vis. Sci. 21, 700-713. Horrocks L. A. (1972) Ether phospholipids. In Ether Lipids (Edited by Snyder), pp. 177-272. Academic Press, New York. Johnston D. and Hudson R. A. (1974) Phospholipids of the cone-rich chicken retina and its photoreceptor outer segment membranes. Biochim. Biophys. Acta 369, 269-277.

Lee T.-C. and Snyder F. (1985) Function, metabolism, and regulation of platelet activating factor and related ether lipids. In Phospholipids and Cellular Regulation. (Edited by Kuo J. F.), Vol. II, pp. l-39. CRC Press, Inc. Boca Raton, FLa. Liang M. T. C., Glonek T., Meneses P., Kopp S. J., Paulson D. J., Gierke L. W. and Schwartz F. N. (1992) NMR spectroscopy study of heart phospholipids. An exercise and anabolic steriods effect. Int. J. Sports Med. 13, 417423.

Liang M. T. C., Meneses P., Glonek T., Kopp S. J., Paulson D. J., Schwartz F. N. and Gierke L. W. (1993) Effects of exercise training and anabolic steroids on plantaris and soleus phospholipids: a )iP nuclear magnetic resonance study. Int. J. Biochem. Cell Biol. 25, 337-347. Mark V., Dungan C. H., Crutchfield M. M. and Van Wazer J. R. (1967) Compilation of P3’ NMR data. Top Phosphor.

Chem.

5, 227458.

Meneses P. and Glonek T. (1988) High resolution “P NMR of extracted phospholipids. J. Lipid Res. 29, 679-689. Meneses P., Navarro J. N. and Glonek T. (1993) Algal phospholipids by “P NMR: comparing isopropanol pretreatment with simple chloroform/methanol extraction. Int.

J. Biochem.

Cell Biol. 25, 903-910.

Meneses P., Greiner J. V. and Glonek T. (1990) Comparison of membrane phospholipids of the rabbit and pig crystalline lens. Exp. Eye Res. 50, 235-240. Meneses P., Para P. F. and Glonek T. (1989) “P NMR of tissue phospholipids: a comparison of three tissue pre-treatment procedures. J. Lipid Res. 30, 458461. Merchant T. E., de Graaf P. W., Minsky B. D., Obertop H. and Glonek T. (1993) Esophageal cancer phospholipid characterization by “P NMR. NMR Biomed. 6, 187-193. Merchant T. E. and Glonek T. (1990) “P NMR of phospholipid glycerol phosphodiester residues. J. Lipid Res. 31, 479-486.

Merchant T. E., Kasimos J. N., de Graaf P. W., Minsky B. D., Gierke L. W. and Glonek T. (199la)

Phospholipid profiles of human colon cancer using “P magnetic resonance spectroscopy. Inr. J. Colorect. Dis. 6, 12ll126. Merchant T. E., Lass J. H., Meneses P., Greiner J. V. and Glonek T. (199lb) Human crystalline lens phospholipid analysis with age. Invest. Ophthalmol. Vis. Sri. 32, 549-55s.

Merchant T. E.. Meneses P., Gierke L. W., Den Otter W. and Glonek T. (1991~) “P magnetic resonance phospholipid profiles of neoplastic human breast tissue. Br. J. Cancer

63, 693-698.

Merchant T. E., Lass J. H., Roat M. I., Skelnik D. L. and Glonek T. (1990) P-31 NMR analysis of phospholipids from cultured human cornea1 epithelial tibroblast and endothelial cells. Curr. Eye Res. 9, 1167-l 176. Nie N. H., Hyadlai Hull C., Jenkins J. G., Steinbrenner K. and Bent D. H. (1975) SPSS: Statistical Package for the SocialSciences, 2nd edn, pp. 181-202. McGraw-Hill, Inc., New York. Philipson K. D., Ders D. M. and Nishimoto A. Y. (1980) The role of phospholipids in the Ca binding of isolated cardiac sarcolemma. J. Molec. Cell Cardiol. 12, 1159~1173. Rietveld A., Berkhout T. A., Roenhorst A., Marsh D. and de Kruijff B. (1986) Preferential association of apocytochrome c with negatively charged phospholipids in mixed model membranes. Biochim. Biophys. Acta 858, 3846.

Rothman J. E. and Lenard J. (1977) Membrane Asymmetry. Science

195, 743-753.

Sachedina S., Greiner J. V. and Glonek T. (1991) Membrane phosphohpids of the ocular tunica fibrosa. Invest. Ophthalmol.

Vis. Sci. 32, 625-623.

Seijo L. M., Merchant T. E., van der Ven L. T. M., Minsky B. D. and Glonek T. (1994) Meningioma phospholipid profiles measured by 3’P magnetic resonance spectroscopy. Lipids 29, 359-364. Smaal E. B., Mandersloot J. G., de Kruijff B. and de Gier J. (1986) Consequences of the interaction of calcium with dioleoylphosphatidate-containing model membranes: changes in membrane permeability. Biochim. Biophys. Acta

860, 99-108.

Tebby J. C. and Glonek T. (1991) “P NMR data of four coordinate phosphorus compounds containing a P==Ch bond but no bonds to H or group IV atoms. Sections H3 and H4: biologically important compounds. In CRC Handbook of Phosphorus-3I Data. (Edited by Tebby),

Boca Raton, Urban P. F., Gangliosides retina. Exp.

Nuclear

Magnetic

Resonance

pp. 262-285. CRC Press, Inc.,

Fla. Edel-Harth S. and Dreyfus H. L. (1975) and phospholipids from frog and duck Eye Res. 20, 397405.