Graefe’s Arch Clin Exp Ophthalmol(2006) 244: 683–688 CLINICAL INVESTIGATION

DOI 10.1007/s00417-005-0141-1

Roselie M. H. DiederenEllen C. La HeijNicolaas E. P. DeutzAlfons G. H. KesselsHans M. H. van EijkFred Hendrikse

Received: 19 July 2005Revised: 30 August 2005Accepted: 30 August 2005Published online: 3 November 2005# Springer-Verlag 2005

Increased nitric oxide (NO) pathwaymetabolites in the vitreous fluid of patientswith rhegmatogenous retinal detachmentor diabetic traction retinal detachment

Abstract Background: Nitric oxide(NO) plays a significant role inphysiological and pathological pro-cesses in the retina. In the L-arginine-NO pathway, NO synthase (NOS)converts L-arginine to NO andL-citrulline. Increased NO production,mediated by inducible NOS has beenimplicated in the pathogenesis ofvarious vitreoretinal diseases. In thepresent study it is hypothesized that inrhegmatogenous retinal detachment(RRD), the production of NO path-way metabolites might be upregu-lated. Methods: Using high-pressureliquid chromatography citrulline, ar-ginine and nitrite were measured invitreous fluid of 93 eyes with RRD,nine eyes with a traction retinaldetachment due to proliferative dia-betic retinopathy (PDR), and in 49control samples of vitreous fluid fromeyes without retinal detachment.Results: The mean vitreous concen-

trations of citrulline and arginine weresignificantly increased in eyes withRRD (9.6±4.3 and 97.3±29.2; re-spectively) or in eyes with a tractionretinal detachment (25.8±10.3 and130.7±23.7; respectively) as com-pared to control eyes (7.1±3.2 and75.9±18.1; respectively). The meanlevel of nitrite was also higher invitreous fluid of patients with RRD(2.24±1.4) or patients with a tractionretinal detachment (2.21±0.72) than inthe controls (2.01±0.72), although notsignificantly so. Conclusions: Wefound increased levels of NO pathwaymetabolites in the vitreous fluid ofeyes with retinal detachment, whichmay reflect a possible role of NO inthe pathogenesis of this disease.

Keywords Retinal detachment .Nitric oxide . Amino acids .Vitreous fluid

Introduction

Nitric oxide (NO) is an important mediator of homeostaticprocesses in the eye, such as regulation of aqueous humourdynamics, retinal neurotransmission and phototransduc-tion [2]. NO is produced when the semi-essential aminoacid L-arginine is converted to L-citrulline and NO by thefamily of nitric oxide synthase (NOS) enzymes [15]. NOthen diffuses into nearby target cells to elevate cyclic GMPlevels and thereby triggers different cell functions.

NOS exists in three major isoforms: neuronal, endothe-lial and inducible NOS. In the retina, neuronal NOS isthought to be responsible for producing NO in photore-

ceptors and bipolar cells, whereas inducible NOS, is foundin Müller cells and in the retinal pigment epithelial (RPE)cell layer [22, 23].

Excess NO may lead to toxic free radical formation,which produces cell death by causing DNA damage. It mayalso cause decreased energy production by inhibition ofmitochondrial function [4]. Inducible NOS is activated byendotoxin and cytokines, while neuronal and endothelialNOS can be activated by glutamate and vasodilatators [2].Over-production of NO may occur when glutamate is gen-erated in excess or when the glutamate re-uptake is inef-fective, as may happen in ischemia. Once glutamate isreleased, it binds to NMDA receptors, increasing the intra-

R. M. H. Diederen (*) . E. C. La Heij .F. HendrikseDepartment of Ophthalmology,University Hospital Maastricht,P. Debyelaan 25,P.O. Box 5800, Maastricht,6202 AZ, The Netherlandse-mail: r.diederen@np.unimaas.nlTel.: +31-43-3875646Fax: +31-43-3875343

N. E. P. Deutz . H. M. H. van EijkDepartment of Surgery,University of Maastricht,Maastricht, The Netherlands

A. G. H. KesselsDepartment of Clinical Epidemiologyand Medical Technology Assessment,University Hospital Maastricht,Maastricht, The Netherlands

cellular calcium concentration and stimulating NOS to pro-duce NO [3].

Thus, excessive glutamate release can cause neurotox-icity via overproduction of NO [4]. High levels of NO havebeen implicated in the pathogenesis of a variety of dis-orders, including glaucoma, proliferative diabetic retinop-athy, cataract and uveitis [7, 8, 17, 20]. However no clinicalstudy has been done previously to investigate whether NOis also involved during retinal detachment.

In a previous study, we found retinal detachment to beassociated with increased levels of vitreous glutamate. Inthe current study, our objective was to find out whetherretinal detachment also leads to a locally increased NOproduction and changes in the concentrations of arginineand citrulline.

Vitreous fluid levels of citrulline, arginine and nitrite,the stable product of NO, were measured using the highlyselective HPLC fluorescence method. Elevated concentra-tions of arginine, citrulline and nitrite were observed in thevitreous fluid of eyes following retinal detachment.

Materials and methods

Patients

In a prospective study, vitreous fluid samples were col-lected from patients with a rhegmatogenous retinal de-tachment (RRD) and from eyes with a traction retinaldetachment due to proliferative diabetic retinopathy (PDR).

All patients were operated on in our department betweenMay 1999 and January 2003 with a vitrectomy. Eyes withuveitis, trauma or macroscopic vitreous haemorrhage wereexcluded. Only patients with a minimum follow-up of3 months were included in the analysis. We operate on eyeswith RRD with up to proliferative vitreoretinopathy (PVR)grade C1 using a conventional scleral buckling technique.Eyes with PVR grade C2 and higher are operated on with aprimary vitrectomy, as described earlier [12]. Vitreoussamples from patients with macular hole or idiopathicepiretinal membranes were used as controls. The study wasperformed with the agreement of the University Hospitalethics committee; all patients gave their informed consentprior to inclusion in the study and after the nature of thestudy had been explained. The study adhered to the tenetsof the Declaration of Helsinki.

The following pre-operative clinical characteristics ofthe study patients were collected for statistical analysis:age, sex, eye affected, the number of detached quadrants ofthe retina, whether or not the central area of the macula(foveal region) was involved in the detachment, pre-op-erative corrected visual acuity, intraocular pressure (IOP),whether the patients used anti-glaucoma medication andwhether the patient had diabetes mellitus. The numberof prior eye operations, including cataract surgery, scleralbuckling, prior endolaser or cryocoagulation, and follow-

up time were also registered. By carefully questioning thepatient, the approximate length of time between the oc-currence of the RRD and the time of sampling was es-tablished. The following variables related to the vitrectomywere collected: whether intraocular endotamponade withoil or gas was necessary and, whether or not a re-detach-ment occurred. At final follow-up we recorded visual acuityand anatomic result.

Sample collection

Undiluted vitrectomy samples were obtained by a conven-tional three-port closed vitrectomy technique by manualsuction at the beginning of the vitrectomy, before openingthe infusion line of Balanced Salt Solution (BSS, AlconLaboratories, Texas, USA), as described earlier [13]. Thesamples were stored at −80°C until the HPLC analysis.Samples contaminated with blood were excluded from thestudy. The measurements were performed in a maskedfashion, without the technician knowing the clinical historyof the patient.

Amino acid analysis

Two amino acids, arginine and citrulline, were analysed asdescribed earlier [21]. Analysis was performed followingprecolumn derivatisation with o-phthaldehyde (OPA) [21].Samples of 50 μl were centrifuged in a Hereaus Biofuge at50,000 g. Next, 4 μl supernatant was pipetted into a glass300-μl insert, already containing 194 μl water and 2 μlnorvaline solution (500 μmol/l). From this mixture, 5 μlwas mixed automatically with 5 μl of OPA reagent, in-cubated for 2 min and injected in a 150×4.6 mm (i.d.)column filled with Allsphere 3 μm (Alltech, Breda,Netherlands) using a WISP 715 sample processor (Waters,Etten-Leur, Netherlands). Amino acid derivatives wereeluted using 25 mmol/l citric acid buffer pH 6.8 containing3% tetrahydrofuran as the starting solvent, followed bygradient elution using a linear addition to 100% of solventB (same buffer, now containing 40% acetonitril and 5%tetrahydrofuran) within 30 min. Fluorescence was mon-itored with a Jasco Model 820FP fluorescence detector (Band L systems, Zoetermeer, Netherlands). Measurementswere made at an excitation wavelength of 335 nm and anemission wavelength of 440 nm. Data were collected on-line and processed using Turbochrom software (Perkin-Elmer, Oosterhout, Netherlands).

Nitrite analysis

The method is based on the automated reaction of NO�2

with 2,3-diaminonaphtalene (DAN) to 1-(H)-naphthatria-

684

zole derivative, followed by a reversed phase separationwith a cycle time (injection to injection) of 20 min. Only a10-μl sample was required, resulting in a linear fluores-cence response in the range from 0.01–10 μmol/l(R2=0.9989). Samples were first centrifuged at 11,000 gin a Heraeus Biofuge Stratos (Dijkstra, Amsterdam, Nether-lands) for 5 min at 4°C. Aliquots were frozen immediatelyin liquid nitrogen and stored at−80°C until analysis. Beforeanalysis, samples were deproteinized through the additionof acetonitril (100 μl acetonitril for each 50-μl sample) in a1.5-ml vial sealed with a screw cap equipped with a rubberO-ring (Sarstedt, Etten-Leur, Netherlands). The resultingsamples were centrifuged for 10 min at 50,000 g at 4°Cusing the Biofuge Stratos. Next, 30 μl supernatant samplealiquots were dispensed into 300-μl spring-mounted plasticinserts into a Waters model WISP 715 auto sampler. Theanalysis was initiated by the addition of 10 μl of a solutionof 0.05 mg diaminonaphtalene (DAN) in 1 ml HCl 0.6 N tothe sample vials. After 1 min, 10 μl of the reaction mix wasinjected onto a 100×4.6-mm (i.d.) column filled with All-sphere C18 3 μm (Alltech, Breda, Netherlands). The DAN-NO2 peak was chromatographed using a 20 min injectioncycle (run to run). At the start of the gradient a 80/20 (v/v)mixture of solvent A [15 mmol/l ammonium acetate (pH8.0), containing 20% (v/v) acetonitrile] and solvent B [40–60 (v/v) mixture of the ammonium acetate buffer-acetoni-trile] was pumped at a flow rate of 1.0 ml/min at roomtemperature resulting in a system back-pressure of 20.0MPa.

Within 8 min a linear gradient was applied to 100% sol-vent B, in which the DAN-NO2 peak elutes after 7.5 min.Hereafter, the column was kept at 100% solvent B for 1min, after which initial conditions were restored in the fol-lowing minute. Fluorescence was monitored with a JascoModel 820 FP fluorescence detector (B&L Systems,Zoetermeer, Netherlands). Measurements were made at anexcitation wavelength of 365 nm and an emission wave-length of 405 nm. Data were collected on-line and processedusing Totalchrom software (Perkin-Elmer, Oosterhout,Netherlands).

Statistical analysis

For statistical analysis, we consulted a professional stat-istician at our University Hospital (A.G.H. Kessels, co-author).

Levels of all neurotransmitters were compared betweengroups using Student’s t-test. Comparisons for sex, glau-coma and prior eye surgery between patients with RRD andcontrols were performed using the chi-square test. ThePearson correlation test was used to test the associationbetween the vitreous concentration of the five neurotrans-mitters and age, intraocular pressure, the number of quad-rants the retina was detached, whether the patient haddiabetes mellitus, the number of prior eye operations, in-cluding cataract surgery, scleral buckling, prior endolaser

Table 1 Basic clinical and ocular characteristics(RRD rhegmatogenous retinal detachment, PDR proliferative diabetic retinopathy)

RRD (n=93) Controls (n=49) PDR (n=9)

Age (years)a 58.0±15.7 68.8±7.7 50.0±5.0Maleb 58 (62%) 24 (49%) 4 (43%)Femaleb 35 (38%) 25 (51%) 5 (57%)Right eyeb 47 (51%) 22 (45%) 4 (43%)Left eyeb 46 (49%) 27 (55%) 5 (57%)IOP (mmHg)a 14.7±5.9 17.1±6.1 13.0±5.2Diabetesb 3 (3%) 3 (6%) 9 (100%)Glaucomab 7 (8%) 4 (8%) 0Aphakicb 6 (6%) 1 (2%) 0Pseudophakicb 27 (29%) 11 (22%) 0Prior scleral bucklingb 62 (54%) 3 (6%) 0Prior endolaserb 26 (28%) 11 (22%) 3 (43%)Prior cryocoagulationb 36 (39%) 3 (6%) 0Duration of detachment (days)a 43.6±72.1Macular detachmentb 61 (66%)Number of quadrants of retinal detachementa 2.6±1.2Post-operative visual acuitya,c 1.49±0.84Post-operative visual acuitya,c 1.28±0.99Re-detachmentb 32 (34%)Duration of follow-up (months)a 24.3±12.5 16.6±14.1 18.7±10.1aNumbers are noted in mean±SDbNumber (percentage)cSnellen visual acuities were converted to a logarithmic scale (logMAR)

685

or cryocoagulation, the approximate length of time be-tween occurrence of retinal detachment and time of sam-pling, whether a re-detachment occurred, and the anatomicresult.

For statistical analysis, Snellen visual acuities were con-verted to a logarithmic scale (logMAR, i.e. the logarithm ofthe minimal angle of resolution), as described earlier [5].The association between the levels of the various neuro-transmitters, and pre-operative visual acuity was investi-gated with a multiple linear regression analysis using thestatus of the macula as co-variable. The association be-tween the levels of the various neurotransmitters, and finalpost-operative visual acuity was investigated with a mul-tiple linear regression analysis using the preoperative visualacuity as co-variable. Finally these associations were inves-tigated with a multivariate linear regression model adjust-ing for nine variables: pre-operative visual acuity, diabetes,glaucoma, prior scleral buckling, prior endolaser, priorcryocoagulation, duration of detachment, status of the mac-ula and number of quadrants of retinal detachment.

Results

Vitreous fluid samples were collected from 93 eyes (93patients, 35 women and 58 men) with rhegmatogenous ret-inal detachment. The mean age was 58 years (SD ±16 years).Mean follow-up time was 23 months (SD ±13 months). Themean duration of retinal detachment was 44 days (rangingfrom 1 to 365 days; Table 1). Furthermore, vitreous sampleswere obtained from patients with traction retinal detach-ment due to PDR (five women and four men, Table 1). Themean age in this patient group was 50 years (SD ±5 years).The control group consisted of 49 patients (25 women and24 men) with an epiretinal membrane or macular hole. Themean age in this group was 69 years (SD ±8 years). Nosignificant differences were found when comparing pa-tients characteristics between patients and controls. Baselineclinical characteristics of all three groups are summarisedin Table 1.

The mean level of citrulline, arginine and nitrite in thevitreous fluid of eyes with traction retinal detachment dueto PDR (25.8±10.3, 130.7±23.7, 2.21±1.53, respectively)

was significantly higher than in the control patients(P<0.001; P<0.001; P=0.004, respectively; Table 2).

The mean level of citrulline in vitreous fluid of eyes withRRD was 9.6±4.3 μM, which was significantly higher thanin the control group (7.1±3.2 μM, P=0.034; Table 2). Themean vitreous concentration of arginine in eyes with RRDwas 97.3±29.2 μM and was also significantly increasedcompared with the control group (75.9±18.1 μM, P=0.002;Table 2). The mean concentration of nitrite was elevated invitreous fluid of RRD patients (2.24±1.4) compared withcontrols (2.01±0.72) although this was not statisticallysignificant (P=0.058).

In the control group, we found no statistically significantdifference in the vitreous concentrations of arginine, cit-

Table 2 Vitreous fluid amino acid concentration (μM)

Controls(n=49)

Eyes withRRD (n=93)

PvalueA

Eyes withPDR (n=9)

PvalueB

Citrulline 7.1±3.2 9.6±4.3 0.034* 25.8±10.3 0.001*

Arginine 75.9±18.1 97.3±29.2 0.002* 130.7±23.7 0.001*

Nitrite 2.01±0.72 2.24±1.45 0.058 2.21±0.72 0.004*

*Statistically significant, P<0.05AEyes with RRD compared with controlsBEyes with PDR compared with controls

5

6

7

8

9

10

11

12

0 - 3 days 4 - 7 days 8 - 30 days more than 30days

Duration of detachment

Cit

rulli

ne

(uM

)

40

50

60

70

80

90

100

110

0 - 3 days 4 - 7 days 8 - 30 days more than 30days

Duration of detachment

Arg

inin

e (u

M)

11.21.4

1.61.8

22.22.4

2.62.8

3

till 4 days 1 week 1 month over 1 month

Duration of detachment

Nit

rite

(u

M)

A

B

C

Fig. 1a–c Correlation between the arginine, citrulline or nitritelevels (μM) in vitreous fluid of patients with a retinal detachmentand the duration of the detachment. The values plotted represent themean levels of arginine, citrulline or nitrite. a Arginine, b citrulline,c nitrite

686

rulline and nitrite between patients with a macular hole andpatients with an epiretinal membrane.

Of all other clinical variables, such as prior scleral buck-ling, prior endolaser, prior cryocoagulation, duration ofdetachment (Fig. 1), status of the macula and number ofquadrants and retinal detachment, analysed in the presentstudy, none correlated significantly with the concentrationof nitrite or arginine. We did observe a statistically sig-nificant correlation between a high level of vitreous cit-rulline and a lower pre-operative visual acuity in patientswith RRD (P=0.027), using the status of the macula as co-variable. Finally, no significant correlation could be foundbetween citrulline, arginine or nitrite and post-operativevisual acuity, after using multivariate analysis.

Discussion

We found significantly increased levels of the NO pathwaymetabolites arginine, citrulline and nitrite in the vitreousfluid of eyes with traction retinal detachment due to PDR.Elevated NO metabolites were also observed in eyes withRRD, but not as striking as in the PDR group. Thesefindings suggest that NO may play a role in the pathogen-esis of retinal detachment.

Only one earlier study reported elevated nitrite levels inthe vitreous fluid of eyes with traction retinal detachmentdue to PDR [22], but levels of nitrite in that study were notcomparable to our data. This may be due to the fact that theseauthors used a spectrophotometric method based on the Griessreaction while we used the more sensitive HPLC technique.

NO is produced by various cell types in the picomolar tonanomolar range, and has a very short half-life (t½<5 s) inbiological fluids. Therefore, it is difficult to measure itspresence directly. NO2

− and NO3−, the stable products of NO

oxidation, are often analysed instead to estimate the NOlevel in biological fluid [15]. The fluorometric HPLCmethod described in the present study offers high re-producibility and specificity for measuring picomolarlevels of NO�

2 . Also, the sensitivity of this HPLC methodfor NO�

2 analysis is greater than that of the batch fluo-rometric method (detection limit, 10–20 mM), the Griesscolorimetric assay (detection limit 0.1– 1 mM), and theHPLC methods with UV-VIS or conductimetric detection(detection limit, 0.1–0.5 mM) [6].

The majority of NO is synthesised by the inducible iso-form of NOS (iNOS) [2]. It has been shown previously thatiNOS is expressed in Müller cells and RPE cells aftercytokine stimulation [9, 11]. Since increased levels ofcytokines have been found in eyes with RRD [10, 13], itmay be postulated that iNOS is upregulated by these cy-tokines, in the detached retina or vitreous fluid of patientswith a retinal detachment. This may be an explanation forthe increased levels of NO pathway metabolites found inthe current study.

NO has many sites of production and actions in theretina, which are both physiological and pathophysiolog-ical in nature. Earlier studies have shown that excess amountsof NO may become damaging to the retina [16, 18]. NOwas found to be functionally as well as histologically toxicto the rabbit retina [18]. In addition, it has been demon-strated that NO is able to mediate excitotoxic and anoxicdamage in retinal ganglion cells [10]. The pathological roleof NO in retinal cell death was supported by the findingthat non-specific inhibition of NOS might protect the retinafrom ischemic damage [19]. Another pathologic mecha-nism may be that overproduction of NO can cause damageto the retina by perturbing rod outer segment (ROS) mem-brane phagocytosis by RPE cells [1]. This could lead to theaccumulation of ROS debris between the photoreceptorsand RPE cells, and could ultimately result in photoreceptordegeneration [1].

Binding of glutamate to the NMDA receptor increasesthe influx of Ca2+, which stimulates NOS to produce NO.The glutamate content of Müller cells has been found to beincreased and remained elevated for many weeks after ex-perimental retinal detachment [14]. In an earlier study, wefound increased levels of glutamate in the vitreous fluid ofpatients with retinal detachment. Thus, increased levels ofglutamate might also be responsible for the increased levelof NO in the vitreous fluid of patients with retinal detachment.

In conclusion, we found increased levels of the NOpathway metabolites in the vitreous fluid of eyes with ret-inal detachment, which may reflect a possible role for NOin the pathogenesis of this disease.

Acknowledgements The authors thank Aize Kijlstra, Albert T.A.Liem and Suzanne C. Dieudonné of the Eye Research InstituteMaastricht, for their help in the preparation of the manuscript. Thisstudy was supported by the Algemene Nederlandse Vereniging terVoorkoming van Blindheid.

References

1. Becquet F, Courtois Y, Goureau O(1994) Nitric oxide decreases in vitrophagocytosis of photoreceptor outersegments by bovine retinal pigmentedepithelial cells. J Cell Physiol 159:256–262

2. Becquet F, Courtois Y, Goureau O(1997) Nitric oxide in the eye: multi-faceted roles and diverse outcomes.Surv Ophthalmol 42:71–82

3. Choi DW, Rothman SM (1990) Therole of glutamate neurotoxicity in hyp-oxic-ischemic neuronal death. AnnuRev Neurosci 13:171–182

4. Dawson VL, Dawson TM (1996) Nitricoxide neurotoxicity. J Chem Neuroanat10:179–190

5. Ferris FL 3rd, Kassoff A, Bresnick GH,Bailey I (1982) New visual acuitycharts for clinical research. AmJ Ophthalmol 94:91–106

687

6. Gharavi N, El-Kadi AO (2003) Mea-surement of nitric oxide in murineHepatoma Hepa1c1c7 cells by reversedphase HPLC with fluorescence detec-tion. J Pharm Pharm Sci 6:302–307

7. Goldstein IM, Ostwald P, Roth S(1996) Nitric oxide: a review of its rolein retinal function and disease. VisionRes 36:2979–2994

8. Inomata M, Hayashi M, Shumiya S,Kawashima S, Ito Y (2000) Amino-guanidine-treatment results in theinhibition of lens opacification andcalpain-mediated proteolysis in Shu-miya cataract rats (SCR). J Biochem(Tokyo) 128:771–776

9. Kelly MEM, Barnes, S (1997)Physiology and pathophysiology ofnitric oxide in the retina. Neuroscientist3:357–360

10. Kenarova B, Voinov L, Apostolov C,Vladimirova R, Misheva A (1997)Levels of some cytokines in subretinalfluid in proliferative vitreoretinopathyand rhegmatogenous retinal detach-ment. Eur J Ophthalmol 7:64–67

11. Koga T, Zhang WY, Gotoh T,Oyadomari S, Tanihara H, Mori M(2003) Induction of citrulline-nitricoxide (NO) cycle enzymes and NOproduction in immunostimulated ratRPE-J cells. Exp Eye Res 76:15–21

12. La Heij EC, Hendrikse F, Kessels AG(2001) Results and complications oftemporary silicone oil tamponade inpatients with complicated retinal de-tachments. Retina 21:107–114

13. La Heij EC, Van De Waarenburg MP,Blaauwgeers HG, Kessels AG, DeVente J, Liem AT, Steinbusch H,Hendrikse F (2001) Levels of basicfibroblast growth factor, glutaminesynthetase, and interleukin-6 in subret-inal fluid from patients with retinaldetachment. Am J Ophthalmol132:544–550

14. Marc RE, Murry RF, Fisher SK,Linberg KA, Lewis GP (1998) Aminoacid signatures in the detached catretina. Invest Ophthalmol Vis Sci39:1694–1702

15. Moncada S, Palmer RM, Higgs EA(1991) Nitric oxide: physiology,pathophysiology, and pharmacology.Pharmacol Rev 43:109–142

16. Morgan J, Caprioli J, Koseki Y (1999)Nitric oxide mediates excitotoxic andanoxic damage in rat retinal ganglioncells cocultured with astroglia. ArchOphthalmol 117:1524–1529

17. Neufeld AH, Sawada A, Becker B(1999) Inhibition of nitric-oxide syn-thase 2 by aminoguanidine providesneuroprotection of retinal ganglioncells in a rat model of chronic glauco-ma. Proc Natl Acad Sci USA 96:9944–9948

18. Oku H, Yamaguchi H, Sugiyama T,Kojima S, Ota M, Azuma I (1997)Retinal toxicity of nitric oxide releasedby administration of a nitric oxidedonor in the albino rabbit. InvestOphthalmol Vis Sci 38:2540–2544

19. Roth S (1997) Role of nitric oxide inretinal cell death. Clin Neurosci 4:216–223

20. Tilton RG, Chang K, Corbett JA,Misko TP, Currie MG, Bora NS,Kaplan HJ, Williamson JR (1994)Endotoxin-induced uveitis in the rat isattenuated by inhibition of nitric oxideproduction. Invest Ophthalmol Vis Sci35:3278–3288

21. Van Eijk HMH, Rooyakkers DR, DeutzNEP (1993) Rapid routine determina-tion of amino acids in plasma by high-performance liquid chromatographywith a 2–3 μm Spherisorb ODS IIcolumn. J Chromatogr 620:143–148

22. Yilmaz G, Esser P, Kociok N, Aydin P,Heimann K, Kociek N (2000) Elevatedvitreous nitric oxide levels in patientswith proliferative diabetic retinopathy.Am J Ophthalmol 130:87–90

23. Yoshida A, Pozdnyakov N, Dang L,Orselli SM, Reddy VN, Sitaramayya A(1995) Nitric oxide synthesis in retinalphotoreceptor cells. Vis Neurosci12:493–500

688