Lactate-Induced Retinal Arteriolar Vasodilation Implicates Neuronal Nitric Oxide Synthesis in Minipigs

Abstract

purpose. To investigate the role of neuronal nitric oxide (NO) synthesis in the retinal vasodilatory response to lactate in minipigs.

methods. Thirteen eyes of 13 minipigs were evaluated. Ten eyes received an intravenous infusion of N ^ω-nitro-l-arginine methyl ester (l-NAME). After 1 hour, the same eyes received an intravitreous juxta-arteriolar microinjection of 30 μL of l-lactate 0.5 M (pH 7.4) through a micropipette. Ten minutes later, 9 of 10 eyes received an intravitreous juxta-arteriolar microinjection of 30 μL of l-NAME 0.01 M (pH 7.4), and 1 received physiologic saline solution (PSS). The remaining three eyes received a microinjection of 30 μL of l-lactate 0.5 M (pH 7.4), without intravenous or intravitreous l-NAME.

results. The three eyes that received juxta-arteriolar injection of l-lactate only showed a reproducible increase in retinal arteriolar diameter that persisted during the entire study period (maximum effect at 20 minutes, 40.9% ± 3.2%). Retinal arteriolar diameter decreased by 4.1% 1 hour after intravenous l-NAME when compared with baseline but the difference did not reach significance. The juxta-arteriolar injection of l-lactate induced a significant increase in retinal arteriolar diameter (22.7% and 28.7% at 5 and 10 minutes, respectively; P < 0.01), followed by a significant decrease (8.6%; P < 0.01) 10 minutes after juxta-arteriolar injection of l-NAME. Injection of PSS had no effect on retinal arteriolar diameter.

conclusions. Juxta-arteriolar administration of l-lactate induced vasodilation, which was also observed with continuous intravenous infusion of l-NAME. Moreover, juxta-arteriolar l-NAME microinjection significantly suppressed the vasodilatory effect of l-lactate. These data suggest that neuronal-derived NO is an important mediator of lactate-induced vasodilation in minipigs.

Achange in retinal arteriolar diameter in response to vasoactive metabolites is essential in the regulation of retinal blood flow to meet the metabolic demands of the retina. Lactate is the major metabolite produced by anaerobic glycolysis in most tissues, but it is also produced under aerobic conditions in the retina. ¹ Animal and human studies have shown that lactate increases retinal blood flow ^{2 3 4} and induces dilation of retinal arteries. ^{5 6 7} However, the possible underlying mechanisms that mediate lactate-induced vasodilation have not been clearly elucidated. Recently, nitric oxide (NO) has been suggested to be the major mediator that causes lactate-dependent vasodilation in isolated porcine retinal arterioles. ⁶ Prostaglandins do not seem to be involved in the vasodilatory response of lactate. ^{5 8}  

NO induces vasodilation through action on vascular smooth muscle cells but may be synthesized at sites other than the endothelium. In the brain, alternate sites of NO production include perivascular nerve fibers and astrocytes adjacent to cerebral blood vessels. ^{9 10 11 12 13 14} Neuronal activity is an important factor in the regulation of cerebral blood flow. In vivo, N-methyl-d-aspartate (NMDA)-induced vasodilatation is attenuated by NO synthase (NOS) blockage, suggesting that vasodilatation is mediated by neuronal rather than endothelial-derived NO. ¹⁵  

Several investigators have examined the role of NO and the endothelium in mediating the vascular tone of isolated artery segments. Relaxation of isolated bovine posterior ciliary arteries was decreased by NOS blockage but was not altered by removal of endothelium, suggesting that release of NO from perivascular nerves is responsible for vasodilation. ¹⁶ In support of this hypothesis, nerve bundles containing NOS have been detected in isolated canine ophthalmic arteries ¹⁷ and in isolated monkey central retinal arteries. ¹⁸ These findings argue that NO acts as a neurotransmitter and that arteriolar muscle tone is regulated by vasodilator nerve activity. 

In the retinal tissue, NOS is found not only in endothelial cells but also in retinal pigment epithelial cells, amacrine cells, horizontal and ganglion cells, nerve fibers in the outer and inner plexiform layers, astrocytes, and Müller cells. ^{19 20} Moreover, retinal neurons and glial cells are in close contact with the vascular wall, and they could thus regulate endothelial cell activity and local blood flow. ^{21 22}  

We tested the hypothesis that lactate-induced vasodilation of retinal arterioles implies neuronal-derived NO in minipigs. Retinal arteriolar diameter changes were measured with a retinal vessel analyzer (RVA; Carl Zeiss Meditec, Jena, Germany). 

Materials and Methods

Experiments were conducted in one eye of 10 minipigs (Göttingen breed, Arare Animal Facility, Geneva, Switzerland) weighing 10 to 12 kg. Minipig retina is an experimental model close to the human retina in both neuroanatomic and vascular aspects. ^{23 24} All the experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Animal Preparation

Minipigs were prepared for the experiments as previously described. ²⁵ After premedication with intramuscular injection of 3 mL (15 mg) of the tranquilizer midazolam maleate (Dormicum; Roche Pharma, Reinach, Switzerland), 3 mL (120 mg) of the tranquilizer azaperone (Stresnil; Janssen Pharmaceutica, Beerse, Belgium), and 1 mL (0.5 mg) atropine, anesthesia was induced with 2 to 3 mg ketamine hydrochloride (Ketalar; Parke-Davis, Zurich, Switzerland) injected into an ear vein. Analgesia was induced with 2 mL (100 μg) of fentanyl (Sintenyl; Sintetica SA, Mendrisio, Switzerland), and curarization was performed with 2 mL (4 mg) of pancuronium bromide (Pavulon; Organon SA, Pfäffikon, Switzerland). The animals were intubated and artificially ventilated. After arterial, venous, and bladder catheterization, anesthesia, analgesia, and myorelaxation were maintained throughout the experiment by continuous perfusion of ketamine, fentanyl, and pancuronium, respectively. 

Each animal was ventilated at approximately 18 strokes/min, with a continuous flow of 20% O₂ and 80% N₂O, through a variable-volume respirator (Sulla 909 V; Dräger, Lübeck, Germany). Systolic and diastolic arterial blood pressure was monitored through the femoral artery with a transducer (Minograph; Siemens-Elema, Göteborg, Sweden). Temperature was maintained between 36°C and 37°C with a thermoblanket. Arterial oxygen partial pressure (Pao ₂), carbon dioxide pressure (Paco ₂), and pH were measured from the same artery with a blood gas analyzer (Labor-system; Flukiger AG, Menziken, Switzerland) and kept under control throughout the experiment by adjusting ventilatory rate, stroke volume, and composition of the inhaled gas. 

A head-holder was used to avoid movements from respiration. The upper and lower eyelids were removed as well as a rectangular area of skin surrounding the eye; the bulbar conjunctiva was detached; the sclera was carefully cleaned to 5 mm from the limbus; the superficial scleral vessels were thermocauterized; the globe was fixed with a metal ring sutured around the limbus; and a sclerotomy was performed 2 to 3 mm posterior to the limbus. A small contact lens with a flat exterior surface was placed on the cornea. The pupil was dilated with 1% atropine eye drops, and the fundus was observed with an operating microscope (Carl Zeiss Meditec, GmbH, Oberkochen, Germany). 

Experimental Procedure

With the minipigs under general anesthesia, 13 eyes of 13 animals were evaluated. Ten minipigs received an intravenous infusion of N ^ω-nitro-l-arginine methyl ester (l-NAME; Sigma-Aldrich Chemie, GmbH, Deisenhofen, Germany). A bolus of 6 mg/kg over 5 minutes was administered followed by a continuous infusion of 60 μg/kg/min that was maintained during the entire study period. After 1 hour of intravenous infusion of l-NAME, an intravitreous juxta-arteriolar microinjection of 30 μL of l-lactate (Sigma-Aldrich Chemie, GmbH) 0.5 M (pH 7.4) was performed. Ten minutes later, an intravitreous juxta-arteriolar microinjection of 30 μL of l-NAME 0.01 M (pH 7.4) was performed in 9 of 10 eyes, and 1 received physiologic saline solution (PSS). The remaining three eyes received an intravitreous juxta-arteriolar microinjection of 30 μL of l-lactate 0.5 M (pH 7.4) only, without intravenous or intravitreous l-NAME. 

Microinjections were performed through a micropipette (tip diameter, 3–4 μm), pulled from a borosilicate capillary. The micropipette was introduced into the vitreous cavity through the pars plana sclerotomy and placed at a distance of 50 to 100 μm from a retinal arteriole (Fig. 1) . 

All the experiments reported herein were performed in conditions of normoxia, normocapnia, and pH 7.4. Care was taken to avoid significant changes in arterial blood pressure, by adjusting the dose of anesthetics accordingly, so as to exclude any effect of blood pressure variations on the retinal arteriolar diameter. ²⁶  

Retinal Arteriolar Diameter Measurements

Retinal arteriolar diameter changes were measured with a commercially available retinal vessel analyzer (RVA; Carl Zeiss Meditec, Jena, Germany). This instrument enables a fast, noninvasive, and objective evaluation of changes in retinal diameter. ^{27 28 29} It comprises a fundus camera (FF 450; Carl Zeiss Meditec), a video camera, a real-time monitor, and a computer with vessel diameter–analyzing software for the accurate determination of retinal vessel diameter. Retinal vessel diameters are analyzed in real time with a maximum frequency of 50 Hz—that is, every second, a maximum of 25 readings of vessel diameter can be obtained. For this purpose, the fundus is imaged onto the charge-coupled device chip of the video camera. The observed fundus images are digitalized with a frame grabber. In addition, the fundus images can be inspected on the real-time monitor and, if necessary, stored with a video recorder. We recorded each experiment with a fundus camera connected to a high-resolution digital video recorder. After the end of each experiment, we connected the video recorder to the RVA and performed the analysis off-line from the recorded video tapes. 

Because of the absorbance properties of hemoglobin, each blood vessel has a specific transmittance profile. Measurement of retinal vessel diameters is based on adaptive algorithms based on these specific profiles. To select a region of interest, the user defines a rectangle on the screen of the real-time monitor. Thereafter, the measurement of vessel diameters can be started. Retinal vessel diameter is then calculated along the arterial segment, which lies within the rectangle. The software calculates the vessel diameter in arbitrary units (AU), which approximately correspond to micrometers at the retinal plane. The retinal vessel diameters are presented in AU. 

Mean Arterial Blood Pressure Calculation

The mean arterial pressure (MAP) was calculated from the diastolic (BP_dia) and systolic (BP_sys) blood pressure according to the equation: MAP = BP_dia + ⅓(BP_sys− BP_dia). 

Statistical Analysis

Repeated-measures analysis of variance (ANOVA) was used to test differences in retinal arteriolar diameter and MAP over the time course of the experiments. Post hoc comparisons were performed with paired t-tests with the Bonferroni correction for multiple comparisons. For data description, drug-induced retinal diameter changes were expressed as the percentage change from baseline. Results are presented as the mean ± SEM. In all comparisons, P < 0.05 defined statistically significant differences. 

Results

Baseline retinal arteriolar diameter (mean ± SEM) was 195 ± 37 AU (n = 10). One hour after intravenous l-NAME infusion, retinal diameter decreased by 4.1% below baseline to 187 ± 36 AU, but the difference did not reach statistical significance. 

Retinal diameter increased significantly (22.7% and 28.7%), 5 and 10 minutes, respectively, after juxta-arteriolar injection of l-lactate. Retinal diameters were 227 ± 46 AU at 5 minutes and 238 ± 45 AU at 10 minutes (P < 0.01), followed by a significant decrease (8.6%) in diameter 10 minutes after juxta-arteriolar injection of l-NAME. Retinal diameter was now 205 ± 31 AU (P < 0.01). Twenty minutes after the juxta-arteriolar injection of l-NAME, the difference was no longer significant when compared with the retinal diameter just before the microinjection of l-NAME. 

Retinal arteriolar diameters are presented in Table 1in AU. Figure 2illustrates the percentage change (mean ± SEM) in retinal diameter during the time course of the experiments. A typical experiment is shown in Figures 3a and 3b . Injection of PSS into one eye that served as the control did not produce any detectable effect on the retinal diameter, and the retinal arterioles remained dilated under the effect of l-lactate (Figs. 4a 4b) . 

During the first 10 minutes after the onset of l-NAME infusion, a 25% mean increase in MAP was observed, which proved the action of l-NAME. Thereafter, MAP was adjusted near baseline, and no significant change in MAP was noted until the end of each experiment, with the exception of a trend toward an increase after 1 hour of intravenous infusion of l-NAME (P = 0.06; Table 2 ). 

To isolate the effect of lactate on retinal arteriolar diameter, we performed microinjections of l-lactate in three eyes of three minipigs that did not receive l-NAME intravenously or intravitreously. Figure 5illustrates the percentage change (mean ± SEM) of retinal diameter up to 30 minutes after the microinjection of l-lactate. There was a reproducible segmental dilation of the retinal arterioles that was observed shortly after the microinjection and was maintained during the whole period of observation. There was an increase in retinal arteriolar diameter of 27.3% ± 10, 37% ± 5.8, 40.9% ± 3.7%, and 40.7% ± 3.2%, respectively, at 5, 10, 20, and 30 minutes after the microinjection. 

Discussion

The role of lactate in retinal vascular regulation has been demonstrated in human and animal studies. ^{2 3 4 6 7 30} Intravenous administration of lactate increases retinal blood flow. ^{2 3 4 30} Changes in retinal blood flow induced by systemic administration of lactate are mainly due to increased blood velocity rather than to changes in retinal vessel diameter. ^{2 5} However, when preretinal juxta-arteriolar microinjection of l-lactate was performed, a significant and reproducible segmental dilation of retinal arterioles was observed. ⁵ These findings suggest that exposure of endothelial and/or glial cells surrounding the retinal arteriolar wall to lactate is not sufficient to trigger a vasomotor effect, unless lactate reaches these cells from the vitreoretinal side. 

The latter ⁵ and other ^{6 7} studies confirm that lactate induces vasodilation in retinal arterioles, in agreement with the findings of our study. The effect of lactate on retinal vascular tone seems to be independent of extracellular pH. Microinjections of both acid and neutral solutions of l-lactate induced segmental vasodilation of similar amplitude in minipigs, and microinjection of an acid solution of d-lactate caused acidification but no vasodilation. ⁵ Prostaglandins are not the principal mediators of the lactate-induced vasodilator effect. ^{5 6 8} Experimental in vivo and in vitro evidence show respectively that perfusion of the eye with indomethacin, an inhibitor of the cyclooxygenase pathway producing prostaglandins, induces vasoconstriction, except at the site where lactate is injected, ⁵ and only slightly reduces the vasodilatory response to lactate in isolated porcine retinal arterioles. ⁶  

Hein et al. ⁶ investigated the role of NO in mediating the vascular response of lactate as well as its underlying signaling mechanisms. They showed that NO is a potent mediator of lactate as l-NAME almost completely inhibited the vasodilatory response of lactate. Theirs is an important study that reveals NO as the major mediator of the lactate vasomotor effect on retinal vessels. Indeed, lactate-induced vasodilation has been shown to be independent of NO in other vascular beds. ^{31 32 33} Hein et al. studied isolated vessel preparations after removal of neuronal tissue surrounding the retinal arterioles, but their study does not reflect what really happens in in vivo conditions. Animal and human studies provide evidence that NO regulates retinal vascular tone in vivo ^{34 35} and that cells other than the endothelial cells produce and release NO in the inner retina. ³⁵ In minipigs, the existence of a preretinal NO gradient from the vitreoretinal surface toward the vitreous indicates a continuous release of NO by the retinal tissue. Such a gradient is also present at a distance from visible arterioles, supporting the hypothesis that retinal cells other than vascular ones seem to release NO. ³⁵  

Three isoforms of NOS have been identified. ^{19 36} The first, termed NOS-I or nNOS, is found in some neurons of the central and peripheral nervous system as well as in some amacrine cells, horizontal cells, and photoreceptors of the retina. The second, NOS-II or eNOS, is mainly expressed by the vascular endothelium. The third, inducible NOS or NO-III is expressed in many cell types, including retinal pigment epithelial cells, Müller cells, retinal pericytes, and capillary endothelial cells, in response to immunologic and inflammatory stimuli. The eNOS isoform has also been found to be expressed by neuronal/glial cells in the central nervous system (CNS) and the retina. ^{37 38} More important, Müller cells and astrocytes express the eNOS isoform. ²⁰ Furthermore, astrocytes and Müller cells have transporters for lactate, localized on the inner plasma membrane of Müller cells and on glial cell processes, ^{39 40} are extensively coupled together by gap junctions ⁴¹ and their processes as well as those of other neuronal cells are in close contact with the basal lamina covering endothelial cells and pericytes of retinal vessels, ^{21 22} suggesting that NO from a neuronal source may play a role in linking retinal blood flow with metabolism, as has been demonstrated in the brain. ^{11 12 13 15 42 43} This effect may be of particular importance in the retina, where it is known that retinal vessels lack autonomic innervation. ^{44 45}  

In an attempt to distinguish between neuronal and endothelial NO in the vasodilator-response to lactate, we used an in vivo setting where l-lactate was injected through a micropipette close to a retinal arteriole under continuous intravenous infusion of l-NAME followed by an intravitreous injection of l-NAME. l-NAME is considered a nonselective inhibitor of both endothelial and neuronal NOS, but the time of inhibition of NOS at the level of the retina probably depends on specific anatomic relations. Unlike other NOS inhibitors, l-NAME does not interact with a specific transporter at the surface of the cells, but relies only on diffusion to enter the cytosol. ⁴⁶ Therefore, and due to anatomic reasons, when it is injected intravenously, it affects NOS in the vascular endothelial cells first and only later blocks NOS within the neuronal/glial cells after diffusion through the blood–retinal barrier. 

Thus, at the level of the retina, during the first hour of intravenous infusion of l-NAME, NOS within the endothelial cells is expected to be predominantly inhibited and the inhibition is expected to occur earlier than that within the neuronal/glial cells. Indeed, there is experimental evidence from studies in brain tissue, showing that intravenous l-NAME blocks endothelial NO over 30 minutes, as indicated by the maximum effect on MAP, and that brain NOS inhibition occurs later. ⁴⁷ A time dependence exists for inhibition of brain NOS. Iadecola et al. ⁴⁸ found that inhibition of brain NOS activity in rats develops over 1 to 2 hours of intravenous infusion of l-NAME, and Traystman et al. ⁴⁷ also found that >70% of NOS inhibition in the brain is apparent in most animals (cats, dogs, and pigs) by 2 hours after intravenous l-NAME. The relatively slow time course of brain NOS inhibition may reflect, at least in part, that l-NAME is not transported across the blood–brain barrier but enters the brain by diffusion. 

In our experiments, measurements before and 1 hour after the onset of intravenous administration of l-NAME revealed that intravenous l-NAME induced a nonsignificant retinal arteriolar vasoconstriction. Keeping the arterial blood pressure stable eliminated the effect due to pressure variations ²⁶ and isolated the effect due to l-NAME itself, which was not statistically significant. Furthermore, juxta-arteriolar injection of l-lactate increased the retinal arteriolar diameter by 22.7% and 28.7%, 5 and 10 minutes later, similar to previous findings, ⁵ suggesting that intravenous infusion of l-NAME does not affect the retinal vasodilatory response to lactate. We used a bolus of 6 mg/kg of l-NAME over 5 minutes followed by a continuous infusion of 60 μg/kg/min, so that after 1 hour, all animals received approximatively a total dose of 10 mg/kg. It has been reported that administration of doses of l-NAME higher than 10 mg/kg do not further increase brain NOS inhibition in rats and that intravenous l-NAME does not lead to complete inhibition of brain NOS even at the highest doses tested (40 mg/kg). ⁴⁸ Nevertheless, it is possible that higher doses would have blocked more of the NOS activity. Indeed, the degree of NOS inhibition after intravenous l-NAME varies widely among species. Traystman et al. ⁴⁷ found that l-NAME yields a dose-related inhibition of brain NOS activity, producing >70% enzyme inhibition at a dose of 20 mg/kg across the species tested (cats, dogs, and pigs). However, an intravenous dose of l-NAME ≥ 10 mg/kg largely reduced brain NOS activity in many animals and MAP increased in all species and at all doses (10–50 mg/kg). This latter finding suggests that even a dose as low as 10 mg/kg inhibits NOS of vascular origin sufficiently. 

When only l-lactate was injected in the preretinal vitreous (n = 3), there was an increase in retinal arteriolar diameter that was maintained during the 30 minutes of the study period. When juxta-arteriolar l-NAME was microinjected (n = 9), the vasodilatory effect of l-lactate was significantly blunted (8.6%) at 10 minutes. Injection of PSS (n = 1) had no effect on the retinal diameter, and retinal arterioles remained dilated under the effect of l-lactate, despite continuous intravenous infusion of l-NAME. When l-NAME was injected into the preretinal vitreous, we expected that it would diffuse first into astrocytes and Müller cells, because the membrane of these cells form the vitreoretinal interface and because the extracellular space of the retina is not directly accessible to the vitreous fluid, suggesting that astrocytes function as communicating elements between the vasculature and the vitreous body. ²¹ Thus, due to this specific anatomy, juxta-arteriolar l-NAME microinjection is expected to affect predominantly the NOS within the neuronal cells. These data support our hypothesis that NO of neuronal origin is an important mediator of lactate-induced vasodilation in minipigs. 

Our study has several limitations. We used l-NAME to inhibit NOS, which is nonselective and blocks both nNOS and eNOS isoforms. However, the purpose of our study was to show that the source of NO mediating the vasodilatory effect of lactate is mainly the neuronal/glial cells and not the endothelial vascular cells and thus to emphasize the role of neuronal/glial cells to the vascular tone regulation. It was not our intent to address which specific isoform is involved, and we cannot do so because of the nonselectivity of l-NAME. Use of selective agents may allow investigators to evaluate the inhibitory effects of specific NO isoforms and to elucidate their respective roles and contributions to retinal vascular regulation. Recently, Pritchett et al. (IOVS 2008;49:ARVO E-Abstract 5370) infused selective and nonselective NOS inhibitors at the vitreoretinal interface in rats and showed that, in vivo, the strongest inhibition of NO production was obtained by infusing 7-nitroindazole, a selective inhibitor of nNOS. They concluded, in support of our hypothesis, that inhibition of nNOS that is found in neuronal/glial cells but not in endothelial cells, appears to play the most significant role in decreasing NO production in vivo. Another limitation of our study is that we did not determine the dose–response characteristics and temporal profile of NOS inhibition in our animal model after intravenous l-NAME. We used relatively low doses of intravenous l-NAME, and we cannot exclude that higher doses would have resulted in more NOS inhibition within the endothelial cells and thus significant vasoconstriction. Given the considerable interanimal variability in response to systemic inhibition, measurement of NOS activity and documentation of adequate inhibition should be considered as a part of the experimental protocol. It should also be noted that part of our experimental protocol, with reference to the dosage and timing of NOS inhibition within the vascular endothelial and the neuronal/glial cells, was based on findings in the brain. Given the anatomic and functional similarities between the two, we extrapolated these findings to the retina. Studies investigating more specifically these parameters in the retina are warranted. 

In summary, we found that juxta-arteriolar administration of l-lactate in minipigs induced vasodilation of retinal arterioles that persisted during the whole study period when only l-lactate was used in the experimental protocol. Lactate-induced retinal arteriolar vasodilation was also observed under continuous intravenous infusion of l-NAME, despite inhibition of endothelial-derived NO. We also demonstrated that juxta-arteriolar l-NAME microinjection inhibiting NOS within the neuronal/glial cells significantly suppressed the vasodilatory effect of l-lactate. Injection of PSS had no effect, and retinal arterioles remained dilated under the effect of l-lactate. Our current findings suggest that, in vivo, neuronal/glial cells are the major source of the NO that mediates the retinal arteriolar vasodilatation response to lactate in minipigs. 

 Supported by Grant 3200B0-105809 from the Swiss National Science Foundation.

 Submitted for publication March 27, 2008; revised June 22, 2008; accepted September 9, 2008.

 Disclosure: E. Mendrinos, None; I.K. Petropoulos, None; G. Mangioris, None; D.N. Papadopoulou, None; A.N. Stangos, None; C.J. Pournaras, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Intravitreous juxta-arteriolar injection through a micropipette. Arrow: outline of the micropipette and its tip, close to a retinal arteriole.

Table 1.

 

Retinal Arteriolar Diameter at Baseline and at Different Time Points

Time Course	Retinal Diameter (mean AU ± SEM)
Baseline	195 ± 12
IV l-NAME (1 h)	187 ± 11
IVT l-lactate (5 min)	227 ± 15
IVT l-lactate (10 min)	238 ± 14
IVT l-NAME (2 min)	213 ± 13
IVT l-NAME (5 min)	209 ± 14
IVT l-NAME (10 min)	205 ± 11
IVT l-NAME (20 min)	203 ± 8.5

 IV, intravenous infusion; IVT, intravitreous juxta-arteriolar microinjection.

Figure 2.

 

Percentage change (mean ± SEM) of the retinal arteriolar diameter compared to baseline, measured with an RVA, during the time course of the experiments (n = 10). One hour after intravenous l-NAME infusion, retinal diameter decreased by 4.1% below baseline, but the difference did not reach statistical significance. Retinal diameter increased significantly (22.7% and 28.7% at 5 and 10 minutes, respectively) after juxta-arteriolar injection of l-lactate (P < 0.01). The increase was followed by a significant decrease of 8.6% in diameter 10 minutes after juxta-arteriolar injection of l-NAME (P < 0.01). Twenty minutes after the juxta-arteriolar injection of l-NAME, the difference was no longer significant when compared with the retinal diameter just before the microinjection of l-NAME.

Figure 3.

 

(a) Retinal arteriolar diameter changes during a typical experiment in a minipig. (a A) Fundus photograph at baseline showing the retinal arterioles that were studied. (a B) One hour after intravenous infusion of l-NAME, there was a mild vasoconstriction of 2% in the retinal arterioles compared with baseline, measured with an RVA. This effect is not evident in the photograph. (a C) Ten minutes after a juxta-arteriolar microinjection of l-lactate, there was a significant increase (23%) in the retinal arteriolar diameter (arrows). (a D) Microinjection of l-NAME produced a decrease of 12% in the retinal diameter at 10 minutes, an effect that is clearly seen in the photograph. (b) Retinal arteriolar diameter changes during a typical experiment in a minipig. Data are from original recordings from the RVA (same experiment as in a). Note that there was a slight decrease in retinal arteriolar diameter after 1 hour of intravenous infusion of l-NAME, followed by a marked increase shortly after juxta-arteriolar intravitreous injection of l-lactate, despite continuous intravenous infusion of l-NAME. This vasodilatory effect was blunted after juxta-arteriolar intravitreous injection of l-NAME. y-Axis: vessel diameter in AU.

Figure 4.

 

(a) Retinal arteriolar diameter changes in the control eye of a minipig that received PSS. (a A) Fundus photograph showing a retinal arteriole crossing over a venule at baseline. (a B) One hour after intravenous infusion of l-NAME, there was a moderate constriction (7%) of the retinal arteriole compared with baseline. This effect can be observed more easily at the point indicated by the arrow. (a C) Ten minutes after a juxta-arteriolar microinjection of l-lactate, there was a significant vasodilation (47%) of the retinal arteriolar diameter. There was also vasodilation of smaller arterioles was more clearly visible (short black arrows). (a D) Injection of PSS did not produce a detectable effect on the retinal diameter. Ten minutes later, the retinal arteriole remained dilated under the effect of l-lactate. (b) Retinal arteriolar diameter changes in the control eye of a minipig that received PSS. Data represent original recordings from the RVA (same experiment as in a). Note that the lactate-induced vasodilation was maintained throughout the whole study period, with injection of PSS having no effect on retinal arteriolar diameter. Intravenous infusion of l-NAME was maintained until the end of the experiment. y-Axis: retinal vessel diameter in AU.

Table 2.

 

MAP at Baseline and at Different Time Points

Time Course	MAP (mean mm Hg ± SEM)
Baseline	57 ± 3
IV l-NAME (1 h)	60 ± 2
IVT l-lactate (10 min)	59 ± 3
IVT l-NAME (10 min)	58 ± 3
IVT l-NAME (20 min)	58 ± 3

 Abbreviations as in Table 1 .

Figure 5.

 

Percentage change (mean ± SEM) in retinal arteriolar diameter compared with baseline, measured with an RVA, after intravitreous juxta-arteriolar injection of l-lactate only (n = 3). There was an increase in retinal arteriolar diameter of 27.3% ± 10%, 37% ± 5.8%, 40.9% ± 3.7%, and 40.7% ± 3.2%, respectively, at 5, 10, 20, and 30 minutes after the injection. These eyes did not receive intravenous injections or intravitreous l-NAME. Note that the retinal arteriolar diameter continued to increase under the effect of l-lactate between 10 and 20 minutes after the injection and was maintained at 30 minutes. This result is contrary to the effect when intravitreous l-NAME was injected at 10 minutes after the injection of l-lactate, reversing the vasodilatory effect of the latter (Fig. 2) .

The authors thank Nicole Gilodi for technical assistance during the animal experiments.