Regulation of the Early Subnormal Retinal Oxygenation Response in Experimental Diabetes by Inducible Nitric Oxide Synthase

The animals were treated in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Association for Research in Vision and Ophthalmology statement on animals in vision research.

Two experimental models of diabetic retinopathy were studied: the diabetic rat and mouse. Each diabetic group started with 15–20 animals. Animals were fed normal rat chow (5001; Ralston Purina, Richmond, IN) and water ad libitum. Rats were maintained in three groups for 3 months: controls (CR; n = 9), untreated diabetes (DR; n = 5), and diabetes treated with the prodrug l-N(6)-(1-iminoethyl)lysine 5-tetrazole amide (12) (100 ppm; D+PRO; n = 6). Prodrug was administered in the chow (i.e., 100 mg compound/kg of chow). The average daily dose depends on the amount of chow consumed (e.g., a 200-g rat consuming ∼20 g of chow per day will receive a dose of ∼10 mg/kg body wt).

Diabetes was induced in rats (starting weight 200–220 g) with an intraperitoneal injection of streptozotocin (55 mg/kg, 0.01 mol/l citrate buffer, pH 4.5) after a 24-h fast and verified 24 h later by the presence of plasma hyperglycemia and glucosuria in nonfasted rats. Rat body weight, average food consumption, and blood glucose levels were monitored weekly. Subtherapeutic levels of insulin (0–4 units of neutral protamine Hagedorn [NPH] insulin administered subcutaneously up to 7 days per week, as needed) were administered to allow slow weight gain, yet maintain hyperglycemia and glucosuria. The diabetic animals produced in this study had mean plasma glucose levels of >400 but <550 mg/dl (Table 1).

Control (C57BL/6J [CM]) and iNOS null (C57BL/6-Nos2^tm1Lau [KO]) mice were purchased from the Jackson Laboratories. The iNOS null mice are viable fertile, nonhypertensive, and have been routinely studied for up to 6 months (V. Laubach, personal communication) (13). Diabetes was induced in C57BL/6J (DM) and in iNOS null (D+KO) mice (starting weight 24–26 g) with streptozotocin (55 mg/kg, 0.01 mol/l citrate buffer, pH 4.5, i.p.) injection once a day for 5 consecutive days. Each day, the food was removed at 8:00 a.m., the animals were injected with streptozotocin at 4:00 p.m., and the food returned. Normal rodent chow and water were provided ad libitum. Body weight was monitored daily and blood glucose levels weekly. Low-dose insulin was administered as needed to maintain a slow weight gain while maintaining the diabetic state in the mouse. After 4 months of diabetes, only those mice with blood glucose levels that were maintained between 400 and 550 mg/dl were studied by MRI.

In CR (n = 7 rats), DR (3 months, n = 6 rats), and D+PRO (3 months, n = 6 rats), both eyes were enucleated and retinas rapidly removed from urethane-anesthetized animals and frozen. As previously described, NO levels were indirectly determined from the concentration of stable metabolites of NO (nitrate plus nitrite) and measured using a fluorimetric assay kit (Cayman Chemical, Ann Arbor, MI) (4). Briefly, retinal homogenates were first filtered (10 kDa) and then incubated again with nitrate reductase for 2 h. Fluorescence generated by nitrite reaction with 2,3-diaminonaphthalene was measured. In all cases, samples were assayed simultaneously.

On the day of the experiment, anesthesia was induced by a single intraperitoneal injection of urethane (36% solution, 0.083 ml/20 g body wt, prepared fresh daily; Aldrich, Milwaukee, WI). Each animal was gently positioned on an MRI-compatible homemade holder with its nose placed in a plastic nose cone. Animals were allowed to breathe spontaneously during the experiment. To maintain the core temperature, a recirculating heated water blanket was used. Rectal temperature, pulse, and hemoglobin oxygen saturation (data not shown) were continuously monitored while the animal was inside the magnet, as previously described (9).

MRI data were acquired on a 4.7 T system using a two-turn transmit/receive surface coil (1.5-cm diameter) placed over the eye. Images were acquired using an adiabatic spin-echo imaging sequence (repetition time [TR] 1 s, echo time [TE] 22.7 ms [the shortest echo time allowed with this sequence], number of acquisitions [NA] 1, matrix size 128 × 256, slice thickness 1 mm, field of view 32 × 32 mm² [for rats], 18 × 18 mm² [for mice], sweep width 25,000 Hz, and 2 min/image) (14). This resulted in an in-plane resolution of 250 × 125 μm² (rat) and 141 × 70 μm² (mouse). Examining the initial retinal oxygenation response to a hyperoxic challenge requires careful attention to experimental timing and makes high demands on the animals’ physiology compared with steady state measurements. We reason that the retinal circulation is often required to maintain an adequate oxygenation level under different conditions (e.g., standing up versus sitting down), and the hyperoxic challenge acts as an acute “stress” test of the retinal circulation’s ability to oxygenate. Other stress tests (e.g., response to hypoxia and/or reduced blood pressure) can also be envisioned. However, relatively smaller signal intensity changes are expected from these challenges compared with that found during carbogen breathing. This would likely make analysis of the data difficult to detect/interpret. In the present study, we chose to use a carbogen challenge because large signal intensity changes are produced in both rat (11) and mouse (data not shown), allowing for the detection of subtler pathophysiology.

The MRI data were collected as follows. Sagittal localizer images were first collected and used to position a single 1-mm transverse slice through the center of the eye. The 1-mm slice thickness was needed to obtain an adequate signal-to-noise ratio in a 2-min image. This slice thickness resulted in some partial volume averaging so that the final image contained superior and inferior hemiretina with some relatively minor contribution from temporal and nasal hemiretina. A capillary tube (1.5-mm inner diameter) filled with distilled water was used as the external standard (not shown in Fig. 1). It is important to note that steady state (room air) vitreous oxygen tension cannot be measured using this method because many factors affect the preretinal vitreous water signal and its relaxation properties. In other words, simply obtaining an image of the eye during room air breathing alone cannot be used to measure retinal oxygenation.

Data were collected as follows: three images while the animal breathed room air and one image during the inhalation of carbogen. Carbogen inhalation was started at the end of the third image. Animals were returned to room air for 5 min to allow recovery from the inhalation challenge and were removed from the magnet. A second 2-min inhalation challenge was performed outside the magnet with care taken to not alter the spatial relationship between the animal head and nose cone. At exactly 2 min, arterial blood from the descending abdominal aorta was collected as previously described (11). This blood was analyzed for glucose, P_aO_2, P_aCO₂, and pH. Note that this second inhalation challenge (outside the magnet) is needed because it is not feasible to routinely obtain an arterial blood sample from inside the magnet (>40 cm away from the magnet opening) from rats or mice. In all cases, after the blood collection, animals were killed with an intracardiac potassium chloride injection.

To be included in this study, the animal must have demonstrated the following traits. 1) There had to be minimal eye movement during the MRI exam. Movement artifacts (typically seen in the phase-encode direction) will confound interpretation of the vitreous signal intensity changes produced during the hyperoxic challenge. 2) The animal had to have a nongasping respiratory pattern before and after the MRI exam. If the animal is gasping, which occurred <1% of the time, the anesthetic was likely improperly administrated (e.g., not intraperitoneally). This could produce a change in systemic oxygenation unrelated to the retinal changes. 3) Rectal temperatures in the range 35.5–36.5°C had to be observed. Preliminary experiments (data not shown) found a strong association between core temperature and P_aCO₂ and P_aO₂ levels. The effect of this correlation on the precision of the measurements was minimized by using a relatively tight range of temperatures. 4) Finally, P_aO₂ >350 mmHg and P_aCO₂ between 46 and 65 mmHg during the carbogen challenge had to be observed. Previously, we found (15) that arterial oxygen levels >350 mmHg during a hyperoxic challenge were needed to produce a consistently large preretinal vitreous oxygenation response. The range of acceptable arterial carbon dioxide levels lie within the array of values in the literature measured under carbogen-breathing conditions. In addition, tight control over the acceptable blood gas value range is needed to ensure adequate quality control of each sample. Occasionally, the blood gas machine was not able to read a sample (e.g., due to a clot or excessive air in the capillary tube). In this case, the MRI data were also excluded. In general, ΔPO₂ data were collected ∼60 min (rats) or 90 min (mice) after urethane injection to avoid potential errors due to variable time under anesthesia. The above acceptance criteria are needed to critically compare the retinal oxygenation response in these spontaneously breathing normal and sick animals while minimizing systemic differences. Because such tight criteria are used, only ∼50% of the animals that started the study were used in the final analysis. Based on our previous experience in rats, n ≥ 5 is sufficient to draw statistical conclusions.

To correct for any movement in the slice plane, a warp affine image coregistration was performed on each animal using software written in-house. This was done for all of the animals used in the final analysis but was only necessary in about half of them regardless of group (e.g., subtle shifting of the animals’ position occurred during the experiment due to settling on the gauze packing). Because the slice thickness (1 mm) is relatively large compared with the diameter of the eye (∼3 mm), partial volumes will be similar if the eye subtly moves out of the imaging plane, and so the data analysis results were not expected to be substantially affected. After coregistration, the MRI data were transferred to a Power Mac G4 computer and analyzed using the program NIH IMAGE (a freeware program available at http://rsb.info.nih.gov/nih-image). Images obtained during room air breathing were averaged to improve the signal-to-noise ratio. All pixel signal intensities in the average room air image and the 2-min carbogen image were then normalized to the external standard intensity. Signal intensity changes during carbogen breathing were calculated and converted to ΔPO₂ values, on a pixel-by-pixel basis, as follows (11). For each pixel, the fractional signal enhancement E was calculated as E = [S(t) − S₀]/S₀, where S(t) is the pixel signal intensity at time (t) after starting the gas inhalation and S₀ is the control signal intensity (measured from the average of the three control images) at the same pixel spatial location. E values were converted into ΔPO₂ using a theory that has been validated in the rat (16): ΔPO₂ = E/(R₁ × T_k), where R₁ is the oxygen relaxivity (s⁻¹ mmHg⁻¹), and T_k = T_r × exp(−T_r/T₁₀), where T_r is the repetition time and T₁₀ is the T₁ in the absence of oxygen. Using a T_r of 1 s, and assuming a vitreous T₁₀ of 4 s, T_k = 3.52. This T₁₀ value is based on our previous measurement of the proton spin-lattice relaxation time in the rabbit vitreous (4 s), reported values in human vitreous (3.3 s), and in cerebral spinal fluid (4.3 s), which has a similar high water content as vitreous (11,17–19). An R₁ of 2 × 10⁻⁴ s⁻¹ mmHg⁻¹ was used. This R₁ was previously measured in a saline phantom, which is assumed to be a reasonable model of vitreous (98% water) (11). A similar R₁ value was found for plasma, suggesting that relatively low protein levels do not substantially contribute to oxygen relaxivity (20). Note that an E of 0.01 (i.e., a 1% signal intensity change) corresponds to a ΔPO₂ of 14 mmHg. There did not appear to be any significant changes in vitreous T₁₀ or R₁ in the animals of this study, based on the similar inferior hemiretinal oxygenation response found in all of the groups in this study.

The ΔPO₂ parameter image was analyzed as follows. First, the pixel values along a 1-pixel thick line drawn at the boundary of the retina/choroid and vitreous were set to 255 (black). We estimated that the thickness of this line, based on the in-plane resolution (e.g., in the mouse, an 18 × 18 mm² field of view was sampled by 128 × 256 datapoints) of 140 × 70 μm², is ∼100 μm. The values in another 1-pixel thick line drawn in the preretinal vitreous next to the black pixels were then extracted (21). This procedure minimized retinal/choroid pixel values from potentially contaminating (“pixel bleed”) those used in the final analysis and insured that similar preretinal vitreous space was sampled for each animal. In addition, spatial averaging over these 100-μm regions of interest will tend to minimize the contribution from the very local preretinal oxygenation gradients next to the retinal surface (22). In addition, an average ΔPO₂ band was constructed based on the within-group mean for each pixel.

The physiological parameters (i.e., blood gas values, rectal temperatures, and blood glucose data) were normally distributed and are presented as mean ± SE. Comparisons were performed using an unpaired t test. Comparison of retinal ΔPO₂ between control and experimental groups was performed using a generalized estimating equation approach. This method performs a general linear regression analysis using all of the pixels in each subject and accounts for the within-subject correlation between adjacent pixels. In all cases, a P ≤ 0.05 was considered significant.

A summary of the rat and mouse systemic physiology is presented in Tables 1 and 2, respectively. As expected, compared with control rats and mice, all diabetic animals had significantly (P < 0.05) elevated blood glucose levels. There were no significant (P > 0.05) differences in plasma glucose levels between the DR and D+PRO groups or between the DM and −D+KO groups.

A summary of the blood parameters measured during a 2-min hyperoxic challenge and core temperature during the experiment are also presented in Tables 1 and 2. The P_aO₂, pH, and core temperature were not significantly (P > 0.05) different among any of the rat groups (Table 1). Although the P_aCO₂ values were significantly lower (P < 0.05) in the D+PRO group compared with those of the DR groups, this degree of arterial hypercapnia is within the range expected during carbogen breathing. Thus, this difference was not considered to be of physiologic significance. The P_aO₂, P_aCO₂, pH, and core temperature were not significantly (P > 0.05) different among any of the mouse groups (Table 2).

Plasma prodrug levels at midstudy (1.5 month) and at the end of study (3 months) were ∼15 μmol/l, which is above the IC₅₀ for iNOS (∼5 μmol/l) and below the IC₅₀ for endothelial NOS (>100 μmol/l) and neuronal NOS (50 μmol/l). The retinal NO level was evaluated using nitrite and nitrate, the stable metabolites of NO. Compared with CR, the retinal concentration of nitrite plus nitrate in DR was elevated ∼1.5-fold (5.3 ± 0.6 vs. 8.1 ± 1.2 pmol/mg protein, P < 0.05, respectively), and in D+PRO animals (2.7 ± 0.28 pmol/mg protein, P < 0.05), the concentration was significantly less than that of DR. These data are consistent with selective inhibition of iNOS in the D+PRO group.

MRI provides a clear image of rodent ocular anatomy in vivo, including the retina/choroid complex (seen as a white line in the posterior region of the eye) and the nucleus and cortex of the lens (Fig. 1) (7).

No significant (P > 0.05) differences were found in average inferior hemiretinal ΔPO₂ comparing any of the rat or mouse groups with their respective normal, nondiabetic controls (CR 169 ± 6 mmHg, DR 121 ± 6, D+PRO 214 ± 9, CM 129 ± 4, DM 142 ± 6, KO 171 ± 6, and D+KO 147 ± 9; mean ± SE). The mean superior hemiretinal ΔPO₂ of the iNOS knockout mice was not significantly (P > 0.05) different from that of controls (Fig. 2). As expected, the average superior hemiretinal ΔPO₂ of the 3-month diabetic rats and 4-month diabetic mice were significantly (P < 0.05) subnormal (9) (Fig. 2). In treated diabetic rats and in the diabetic iNOS knockout mice, the mean superior hemiretinal ΔPO₂ values were not different (P > 0.05) from that of their respective controls (Figs. 2–4). To examine the regional effect of iNOS inhibition in diabetic rats and mice, we compared plots of the average superior hemiretinal pixel ΔPO₂ values versus pixel position (i.e., distance) from the optic nerve head. As seen in Figs. 3 and 4, there is clear evidence that diabetic iNOS-inhibited rodents experienced a pansuperior retinal correction in ΔPO₂.

In this study, we confirmed that diabetic rats experience an increased production of retinal NO (4) and that a relatively specific inhibitor of iNOS activity [prodrug of l-N(6)-(1-iminoethyl)lysine 5-tetrazole amide] prevented this accumulation of retinal NO. In addition, we found a subnormal superior hemiretinal ΔPO₂ in rat and mouse models of diabetic retinopathy and examined the contribution of iNOS to this diabetes-induced pathophysiology. Phenotyping requires an accurate measurement of the retinal ΔPO₂. Previously, we reported agreement between functional MRI and oxygen electrode ΔPO₂ data in control rats. In addition, no significant differences (P > 0.05) were found in the inferior hemiretinal ΔPO₂ between any of the groups studied. These considerations provide confidence in the assumed T₁₀ and R₁ values used in the second equation and imply that potential artifacts (such as those from data scatter or mixing of vitreous from outside the slice) do not confound data interpretation in the rat or mouse. Retinal oxygen measurements using a microelectrode have not been performed in the mouse, so it is not possible to fully determine the accuracy of the murine functional MRI measurement. However, similar ΔPO₂ values were found in control rats and mice (Fig. 2). Together, these considerations strongly support the accuracy of the functional MRI ΔPO₂ measurement in rats and mice.

The subnormal superior hemiretinal ΔPO₂ found in 3-month diabetic rats and 4-month diabetic mice appears to be an early event in the sequence leading to histopathology in diabetes. Furthermore, treatments that prevented retinal lesion formation in long-term diabetic and galactosemic rats also prevented the development of a subnormal superior hemiretinal ΔPO₂ in more acute studies (9). These observations strongly suggest that subnormal superior hemiretinal ΔPO₂ is an early functional lesion that predicts the development of the structural lesions. The vascular lesions of diabetic retinopathy in rats or mice take longer to develop than the 3- to 4-month duration used in the present study (7,23,24), so we cannot currently evaluate the relation of NO to the development of histopathology. It was beyond the scope of this study to determine why the superior hemiretinal ΔPO₂ becomes subnormal and the inferior hemiretinal does not. Nonetheless, based on the above, it appears that subnormal retinal ΔPO₂ occurs before the appearance of experimental diabetic retinopathy.

Previous studies have shown that in diabetic rats, AMG, which is a relatively selective inhibitor of iNOS activity (among other actions), significantly inhibited diabetes-induced nitrative stress by 2 months, corrected the development of a subnormal superior hemiretinal ΔPO₂ by 3 months, and prevented the development of significant retinal histopathology by 15 months (4,9,25). However, based on these data, the association between diabetes-induced iNOS and the development of a subnormal ΔPO₂ is indirect. The major finding in this study is that the inhibition of iNOS activity, either pharmacologically or genetically, prevented development of a subnormal superior hemiretinal ΔPO₂ in diabetic rodents. One possible implication of the present study is that diabetes induces a greater increase in iNOS activity (and presumably NO level) in the superior hemiretina than in the inferior hemiretina. More work is needed to test this hypothesis.

The exact mechanism(s) by which iNOS produces a subnormal ΔPO₂ remains unclear. In diabetic rats and mice, chronic hyperglycemia produces an elevated retinal iNOS expression resulting in increased NO levels (3,4). Such nitrative stress may produce chronic vasodilatation that, coupled with other complications of diabetes, such as increased advanced glycated end products, could prevent the retinal vessel endothelium from responding appropriately to a carbogen challenge. Experiments are ongoing in the laboratory to further evaluate this speculation. It remains to be determined whether diabetic rodents treated with a specific inhibitor of iNOS activity will have subsequent retinal histopathology. Nonetheless, the present data establish that diabetes-induced increased iNOS activity as a key step in the development of early diabetic pathophysiology.

Support for this work was provided by National Institutes of Health Grants nos. RO1 EY10221 (to B.A.B.), EY00300 (to T.S.K.), by the Research to Prevent Blindness, and by the Juvenile Diabetes Research Foundation (to B.A.B.).

We are grateful to Dr. Tom Hohman for his careful reading of this manuscript.