Inner Retinal Oxygen Delivery and Metabolism in Streptozotocin Diabetic Rats

Abstract

Purpose.

The purpose of the study is to report global measurements of inner retinal oxygen delivery (DO_{2_IR}) and oxygen metabolism (MO_{2_IR}) in streptozotocin (STZ) diabetic rats.

Methods.

Phosphorescence lifetime and blood flow imaging were performed in rats 4 (STZ/4wk; n = 10) and 6 (STZ/6wk; n = 10) weeks following injection of STZ to measure retinal arterial (O_2A) and venous (O_2V) oxygen contents and total retinal blood flow (F). DO_{2_IR} and MO_{2_IR} were calculated from measurements of F and O_2A and of F and the arteriovenous oxygen content difference, respectively. Data in STZ rats were compared to those in healthy control rats (n = 10).

Results.

Measurements of O_2A and O_2V were not significantly different among STZ/4wk, STZ/6wk, and control rats (P ≥ 0.28). Likewise, F was similar among all groups of rats (P = 0.81). DO_{2_IR} measurements were 941 ± 231, 956 ± 232, and 973 ± 243 nL O₂/min in control, STZ/4wk, and STZ/6wk rats, respectively (P = 0.95). MO_{2_IR} measurements were 516 ± 175, 444 ± 103, and 496 ± 84 nL O₂/min in control, STZ/4wk, and STZ/6wk rats, respectively (P = 0.37).

Conclusions.

Global inner retinal oxygen delivery and metabolism were not significantly impaired in STZ rats in early diabetes.

Measurements of inner retinal oxygen delivery and oxygen metabolism in streptozotocin diabetic rats indicate that oxygen supplied by the retinal circulation and consumed by the inner retinal tissue were not altered within 6 weeks of diabetes.

Introduction

Diabetic retinopathy (DR) is a significant cause of blindness in developed countries.^1–3 It is thought that in some stages of DR, the ability of the retinal vasculature to deliver oxygen and of the retinal tissue to consume oxygen may be impaired. Retinal blood flow, as a marker of oxygen delivery, has been studied extensively in DR subjects using a variety of techniques such as fluorescein angiography,^4–7 laser Doppler velocimetry,^8–11 and blue field entoptic phenomenon.^12,13 Findings of these studies have been variable, with reports of reduced, unchanged, or elevated blood flow.^14–18 Due to limited technologies for measuring inner retinal oxygen consumption, abnormalities in retinal vascular oxygenation have been sought as a surrogate for impaired oxygen metabolism. In subjects with DR, retinal venous oxygen saturation (SO₂) measured by oximetry was found to be increased^19,20 while arterial SO₂ was reported to be higher than²⁰ or similar¹⁹ to values in healthy subjects. Combined blood flow and oximetry can be used to estimate inner retinal oxygen metabolism (MO_{2_IR}). However, this combination has been used in only one study, and that was of neurogenic optic atrophy.²¹ Therefore, in humans, the effect of diabetes on MO_{2_IR} remains unknown. Knowledge of alterations in inner retinal oxygen delivery (DO_{2_IR}) and MO_{2_IR} would be useful to better understand DR pathophysiology and potentially for development of therapeutic interventions.

The streptozotocin (STZ) diabetic rat develops retinal pathological alterations that resemble those observed in early human DR.²² In early experimental diabetes (within 6 weeks), blood–retinal barrier breakdown,^23–26 leukostasis,^23,27,28 upregulation of endothelin-1,²⁹ abnormal retinal blood flow,^30–34 and reduced retinal arterial wall oxygen tension (PO₂)³⁵ have been reported in STZ rats, suggesting the possibility of alterations in DO_{2_IR}. However, significant reductions in tissue PO₂ may not occur even if oxygen delivery is reduced to some extent. In fact, indicators of hypoxia measured by pimonidazole³⁶ and hypoxia-inducible factor (HIF) levels³⁷ were not abnormal in excised retinal tissue of early STZ rats, though upregulation of vascular endothelial growth factor (VEGF)^24,38,39 has been reported. In addition to oxygen availability, retinal neural activity is also a determinant of the rate at which the retina consumes oxygen for energy generation. In early STZ rodents, neural changes including increased apoptosis of retinal ganglion cells,^40–43 thinning of the inner plexiform layer,⁴⁴ and reduction of amacrine cells⁴⁵ have been observed before the appearance of vascular cell changes,²² suggesting that there may be alterations in MO_{2_IR}. However, to date, measurements of neither oxygen delivery nor oxygen metabolism in the inner retina in living STZ rats have been published. The purpose of this study was to quantitatively assess DO_{2_IR} and MO_{2_IR} in STZ rats with our previously established optical imaging method.^46,47

Methods

Animals

Male Long Evans pigmented rats (Charles River Laboratories, Wilmington, MA) were treated in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Diabetes was induced by intravenous injection of STZ (60 mg/kg) in citrate buffer. Nonfasting blood glucose levels were measured in STZ rats weekly and immediately before imaging with the use of a commercially available blood glucometer (FreeStyle Lite; Abbott, Alameda, CA) to confirm continued hyperglycemia. Imaging was performed either 4 weeks (STZ/4wk; n = 10) or 6 weeks (STZ/6wk; n = 10) after administration of STZ. Imaging in rats with longer duration of diabetes was precluded due to cataract formation.

Prior to imaging, rats were anesthetized with intraperitoneal injections of ketamine (100 mg/kg) and xylazine (5 mg/kg), with additional injections given to sustain anesthesia as necessary. To ensure normal systemic blood gas levels, rats were mechanically ventilated with room air (or room air and supplemental oxygen) using a small-animal ventilator (Harvard Apparatus, Inc., South Natick, MA) connected to an endotracheal tube. The femoral artery was cannulated, and a catheter was attached to draw blood and to measure the animal's physiological status with a pressure transducer. Systemic arterial oxygen tension (P_aO₂), carbon dioxide tension (P_aCO₂), and pH were measured immediately prior to imaging from arterial blood using a blood gas analyzer (Radiometer, Westlake, OH) 5 to 10 minutes after initiation of ventilation. Blood pressure (BP) and heart rate (HR) were monitored continuously with a data acquisition system (Biopac Systems, Goleta, CA) linked to the pressure transducer. Continuous BP and HR measurements obtained during imaging were averaged to derive a representative BP and HR value. Hemoglobin concentration (HgB) was also measured with a hematology system (Siemens, Tarrytown, NY) from arterial blood.

Rats were placed in an animal holder with a copper tubing water heater to maintain the body temperature at 37°C. Pupils were dilated with 2.5% phenylephrine and 1% tropicamide. A glass cover slip with 1% hydroxypropyl methylcellulose was applied to the cornea to eliminate its refractive power and prevent dehydration. For retinal vascular PO₂ imaging, an oxygen-sensitive molecular probe, Pd-porphine (Frontier Scientific, Logan, UT), was dissolved (12 mg/mL) in bovine serum albumin solution (60 mg/mL) and administered through the femoral arterial catheter (20 mg/kg). For retinal blood velocity imaging, 2-μm polystyrene fluorescent microspheres (Invitrogen, Grand Island, NY) were injected through the catheter. For vascular caliber measurement, red-free retinal imaging was performed; and in three STZ rats with low-quality red-free images, fluorescein angiography (FA) was performed by intravascular injection of 10% fluorescein sodium (5 mg/kg, AK-FLUOR; Akorn, Decatur, IL) for improved visualization of vessel diameter. Overall, the duration of the vascular PO₂ and blood flow imaging session was approximately 1 hour.

Oxygen Tension Imaging

Retinal vascular PO₂ measurements were obtained using our optical section phosphorescence lifetime imaging system.^48,49 Briefly, a laser line was projected on the retina after intravenous injection of the Pd-porphine probe. Due to the angle between the excitation laser beam and imaging path, an optical section phosphorescence image was acquired in which the retinal vessels were depth-resolved from the underlying choroid. Phosphorescence lifetimes in the retinal vessels were determined using a frequency-domain approach and converted to PO₂ measurements using the Stern-Volmer equation.^50,51 PO₂ was measured in all major retinal arteries (PO_2A) and retinal veins (PO_2V) at locations within three optic disc diameters (∼600 μm) from the edge of the optic nerve head. Three repeated PO₂ measurements were averaged per blood vessel.

Blood Flow Imaging

Blood flow was measured using our previously described imaging system.^46,47 Briefly, a slit-lamp biomicroscope, equipped with a 488-nm diode laser (excitation), emission filter (560 ± 60 nm), and a high-speed charge-coupled device camera (QImaging, Surrey, Canada), was utilized for imaging of intravascular motion of the fluorescent microspheres at 108 Hz to determine retinal venous blood velocity (V). Using the same instrument, retinal venous diameter (D) was measured by red-free retinal imaging in 17 STZ rats or by FA in 3 STZ rats that had low-quality red-free images. Red-free retinal images were captured with the instrument's light source and a green filter (540 ± 5 nm), while FA images were obtained by placing a fluorescein excitation filter (488 ± 5 nm) in the illumination path and using the emission filter. Venous D was determined based on the average full width at half maximum of intensity profiles perpendicular to the blood vessel axis over a fixed length of 200 μm. Blood flow in each major vein was then calculated from V and D measurements (V×π×D²/4) and summed over all veins to provide total blood flow (F) in the retinal circulation. Blood flow was determined in major retinal veins because they are less affected by pulsation, and they have higher contrast (darker) and larger diameters compared to major retinal arteries.

Global Inner Retinal Oxygen Delivery and Metabolism

The oxygen content of blood was calculated from PO₂ measurements for each major retinal artery and vein as the sum of oxygen bound to hemoglobin and dissolved in blood, as previously described.^46,47 The amount of oxygen bound to hemoglobin was calculated from the hemoglobin dissociation curve in rat using the measured vascular PO₂, pH, and HgB values,⁵² while the amount of dissolved oxygen was determined from the product of the vascular PO₂ and oxygen solubility in blood. Retinal arterial (O_2A) and venous (O_2V) oxygen contents were determined by averaging the oxygen content of all major retinal arteries and veins, respectively. DO_{2_IR} was derived from the product of F and O_2A, while MO_{2_IR} was calculated from the product of F and the arteriovenous oxygen content difference (O_2A-V).

Data Analysis

Data obtained in STZ/4wk and STZ/6wk rats were compared to our previously published data⁴⁷ in 10 healthy control rats using analysis of variance. Mean ages of control, STZ/4wk, and STZ/6wk rats on the day of imaging were 85 ± 19, 102 ± 3, and 109 ± 12 days, respectively. Post hoc analysis using Tukey's method was performed to determine pairwise differences. Linear regression analysis was also performed to determine if F was significantly related to systemic BP or HR in each group of rats. Statistical significance was accepted at P < 0.05.

Results

Blood Glucose and Systemic Physiological Status

Blood glucose measurements, obtained on the day of imaging, were elevated in all STZ rats (>314 mg/dL), confirming the presence of diabetes. There was a significant difference in body weight among control (391 ± 79 g, mean ± SD, n = 10), STZ/4wk (339 ± 43 g, n = 10), and STZ/6wk (307 ± 34 g, n = 10) rats (P = 0.008), with the only significant pairwise difference between control and STZ/6wk rats (P = 0.006).

The systemic physiological status of control and STZ rats is summarized in Table 1. P_aO₂, P_aCO_2, pH, and HgB were within normal ranges and not significantly different among control, STZ/4wk, and STZ/6wk rats (P ≥ 0.23). However, BP was significantly different among the groups (P < 0.001), with both STZ/4wk and STZ/6wk rats having lower BP compared to controls (P ≤ 0.03). Likewise, HR was significantly different among the groups (P = 0.04), with STZ/6wk rats having lower HR compared to control rats (P = 0.04).

Table 1.

Systemic Physiological Parameters (Mean ± SD) in Control, STZ/4wk, and STZ/6wk Rats

Systemic Physiologic Parameters	Control,* n = 10	STZ/4wk, n = 10	STZ/6wk, n = 10	P Value
Arterial PO2, mm Hg	91 ± 9	85 ± 19	87 ± 14	0.64
Arterial PCO2, mm Hg	42 ± 3	40 ± 4	43 ± 10	0.54
pH	7.36 ± 0.04	7.33 ± 0.08	7.29 ± 0.11	0.23
Blood pressure, mm Hg	101 ± 18	79 ± 23	64 ± 12	<0.001
Heart rate, beats/min	224 ± 35	193 ± 40	183 ± 33	0.04
Hemoglobin concentration, g/dL	14.2 ± 0.8	14.2 ± 1.1	14.2 ± 1.4	0.98

^* Previously published data.⁴⁷

Retinal Vascular PO₂ and Oxygen Content

Mean values of retinal vascular PO₂ and oxygen contents, compiled from measurements in all rats, are summarized in Table 2. Differences in PO_2A measurements among control, STZ/4wk, and STZ/6wk rats approached significance (P = 0.06), and no significant difference was found in PO_2V (P = 0.10). After converting PO₂ measurements to oxygen contents, there were no significant differences in O_2A, O_2V, and O_2A-V among the groups of rats (P ≥ 0.28).

Table 2.

Retinal Arterial (PO_2A) and Venous (PO_2V) Oxygen Tension, Arterial (O_2A) and Venous (O_2V) Oxygen Content, and Arteriovenous Oxygen Content Difference (O_2A-V) of Control, STZ/4wk, and STZ/6wk Rats (Mean ± SD)

Retinal Oxygenation Parameters	Control,* n = 10	STZ/4wk, n = 10	STZ/6wk, n = 10	P Value
PO2A, mm Hg	44 ± 4	49 ± 8	53 ± 11	0.06
PO2V, mm Hg	26 ± 3	30 ± 5	31 ± 8	0.10
O2A, mL O2/dL	11.9 ± 1.4	12.7 ± 2.9	13.0 ± 1.9	0.53
O2V, mL O2/dL	5.4 ± 1.3	6.8 ± 2.3	6.5 ± 2.3	0.28
O2A-V, mL O2/dL	6.5 ± 1.5	5.9 ± 1.2	6.5 ± 1.4	0.49

^* Previously published data.⁴⁷

Retinal Blood Flow

Mean values of D, V, and F, compiled from measurements in all rats, are summarized in Table 3. Venous D and V were not significantly different among control, STZ/4wk, and STZ/6wk rats (P ≥ 0.60). Likewise, F was similar among the groups of rats (P = 0.81). In each group, F was not significantly correlated with either BP (R² ≤ 0.23, P ≥ 0.15, n = 10) or HR (R² ≤ 0.15, P ≥ 0.26, n = 10).

Table 3.

Retinal Venous Diameter (D), Velocity (V), and Blood Flow (F) of Rats (Mean ± SD) in Control, STZ/4wk, and STZ/6wk Rats

Blood Flow Parameters	Control,* n = 10	STZ/4wk, n = 10	STZ/6wk, n = 10	P Value
D, μm	51 ± 6	50 ± 9	52 ± 6	0.85
V, mm/s	11.8 ± 2.8	12.6 ± 2.4	11.5 ± 2.7	0.60
F, μL/min	7.9 ± 1.7	7.7 ± 1.8	7.4 ± 1.1	0.81

^* Previously published data.⁴⁷

Global Inner Retinal Oxygen Delivery and Metabolism

Measurements of DO_{2_IR} and MO_{2_IR} in individual control and STZ rats are shown in the Figure. Mean and SD values of DO_{2_IR}, calculated for each group, were 941 ± 231, 956 ± 232, and 973 ± 243 nL O₂/min in control, STZ/4wk, and STZ/6wk rats, respectively. DO_{2_IR} was similar among control, STZ/4wk, and STZ/6wk rats (P = 0.95). Mean and SD values of MO_{2_IR} were 516 ± 175, 444 ± 103, and 496 ± 84 nL O₂/min in control, STZ/4wk, and STZ/6wk rats, respectively. Likewise, MO_{2_IR} was not significantly different among the groups of rats (P = 0.37).

Figure.

Measurements of (A) inner retinal oxygen delivery (DO_{2_IR}) and (B) inner retinal oxygen metabolism (MO_{2_IR}) in individual control (n = 10), STZ/4wk (n = 10), and STZ/6wk (n = 10) rats. Solid horizontal lines indicate the mean and standard deviation of measurements in each group.

Discussion

Although retinal alterations in early STZ rats have been observed in published studies,^{23–35,38–45} corresponding changes in DO_{2_IR} and MO_{2_IR} have not been previously reported. In the present study, combined measurements of DO_{2_IR} and MO_{2_IR} are reported in living STZ rats 4 and 6 weeks after induction of diabetes, suggesting that DO_{2_IR} and MO_{2_IR} are not significantly altered in early experimental diabetes.

Systemic arterial blood gases of control and STZ rats in the current study were within normal ranges due to controlled ventilation, thereby minimizing the potentially confounding effects of systemic blood gas tensions on retinal blood flow.¹⁷ However, BP and HR were lower in STZ rats compared to healthy control rats, similar to previously reported findings.^53,54 One may have expected reduced BP and HR in STZ rats to affect retinal blood flow. However, F was not correlated with either BP or HR; therefore the decrease of these parameters in STZ rats likely did not significantly influence measurements of DO_{2_IR} and MO_{2_IR}.

Retinal vascular PO₂ in STZ rats after 4 and 6 weeks of diabetes was comparable to that of healthy control rats. In control rats, retinal PO_2A and PO_2V measurements were in general agreement with previously reported values obtained with microelectrodes at the retinal arterial and venous walls in healthy rats, respectively.^35,47,55 Similar to findings of the current study, PO₂ measured at the retinal venous wall was not different between STZ and healthy rats, while contrary to our findings, PO₂ measured at the retinal arterial wall was lower in STZ rats.³⁵ This difference may possibly be attributed to reduced oxygen diffusion across the arterial wall in diabetes, such that the extraluminal PO₂ would be lower than the intraluminal PO₂. The finding of comparable O_2A and O_2V among STZ and healthy control rats is different from reported abnormalities in retinal arterial and venous SO₂ in human subjects with DR.^19,20 This may be attributed, at least in part, to differences in duration of diabetes and severity of retinopathy.

Blood flow and DO_{2_IR} were not altered in STZ rats after 4 or 6 weeks of diabetes as compared to healthy control rats. Retinal blood flow measurements in control rats were comparable to values reported using Doppler OCT⁵⁶ and fluorescent microsphere impaction.⁵⁷ Similar to the findings of the current study, retinal blood flow has been reported to be unchanged in STZ rats at 3 weeks⁵⁸ and STZ mice at 2.5 months,⁵⁹ while other published studies have shown reductions or elevations of blood flow in STZ rodents.^30–34 Inner retinal oxygen delivery in control rats was previously reported by us,⁴⁷ but to our knowledge there have been no other published data in healthy or diabetic animals. The finding of comparable DO_{2_IR} between STZ and healthy rats suggests that previously reported vascular abnormalities^23–29 were not sufficiently severe or widespread within 6 weeks of experimental diabetes to significantly impair retinal hemodynamics.

Inner retinal oxygen metabolism measurements were similar among STZ and healthy control rats in the current study. Oxygen metabolism measurements of whole retinas removed from alloxan-induced diabetic rats were higher than in healthy control retinas,⁶⁰ though glucose metabolism measurements in STZ rats were not elevated by hyperglycemia.⁶¹ Although outer retina oxygen consumption of long-term diabetic cats has been reported,⁶² it is difficult to compare these results to our findings of inner retina oxygen consumption of short-term diabetic rats because of differences in species, duration of diabetes, and retinal cell layers. Since DO_{2_IR} was not reduced due to diabetes, MO_{2_IR} was not limited by oxygen supply. Furthermore, the finding of unaltered MO_{2_IR} in STZ rats suggests that previously reported loss of inner retinal neurons^40–45 may not have significantly impacted the overall neuronal activity in early experimental diabetes.

Since retinal tissue oxygen levels reflect the balance between oxygen delivery and metabolism, the findings of unaltered DO_{2_IR} and MO_{2_IR} in the current study suggest similar retinal tissue PO₂ in STZ and control rats. In agreement with our findings, inner retinal tissue PO₂ measured by oxygen microelectrodes was comparable between 4-week STZ and healthy rats (Lau JC, et al. IOVS 2010;51:ARVO E-Abstract 5644). Similarly, hypoxia indicators using pimonidazole in the retinal tissue were not altered in 3-week STZ rats, though the HIF-2α level was increased in some peripheral retinal layers.³⁶ Furthermore, HIF-1α and HIF-2α levels in the retinal tissue of 4-week STZ rodents were not increased.³⁷ In later stages of experimental diabetes between 3 and 6 months, hypoxia indicators and VEGF expression were abnormal in the retinal tissue of STZ rodents,^37,63 while other studies reported that hypoxia indicators (Reeves D, et al. IOVS 2002;43:ARVO E-Abstract 785) and the ratio of nicotinamide adenine dinucleotide (NAD⁺) to NADH⁶⁴ were not altered.

There were limitations in the current study. Imaging of STZ rats beyond 6 weeks of diabetes was impeded due to cataract formation, thereby prohibiting measurements with our optical methods. Although Pd-porphine has been shown to be phototoxic to the retina under high light levels and long exposures,⁶⁵ given the retinal irradiance levels used for vascular PO₂ imaging, it is unlikely that the results of the current study were affected. The small sample size reduced the statistical power to detect subtle differences among STZ and healthy rats, given the interanimal variability of the measurements. Blood glucose was not measured in control rats, though it was expected to be significantly lower than in STZ rats because the two groups were maintained under identical conditions with the exception of STZ administration. Since sham injections were not administered in control rats, and since control and diabetic rats were not litter matched, these factors may have affected the results. Oxygen delivered by the retinal circulation supplies approximately 7% of the metabolic needs of the photoreceptors under light-adapted conditions⁶⁶; thus a small fraction of the outer retinal oxygen consumption is reflected in the reported MO_{2_IR} measurements. Since DO_{2_IR} and MO_{2_IR} measurements represented global values derived for the inner retina of the whole eye, it is possible that oxygen delivery and consumption in local retinal areas may have been impaired in early experimental diabetes. In future studies, our alternative method⁴⁶ to measure DO₂ and MO₂ in local regions of the retina may be utilized to investigate multifocal abnormalities in STZ rats.

In conclusion, inner retinal oxygen delivery and metabolism were not significantly altered within 6 weeks of experimental diabetes. Findings from this study may imply minimal alterations in global inner retinal oxygen delivery and metabolism in early human diabetes.