Marked variation in responses to long-term nitric oxide inhibition during pregnancy in outbred rats from two different colonies

Efficacy of Dynamics-based Features for Machine Learning Classification of Renal Hemodynamics

Different machine learning approaches for analyzing renal hemodynamics using time series of arterial blood pressure and renal blood flow rate measurements in conscious rats are developed and compared. Particular emphasis is placed on features used for machine learning. The test scenario involves binary classification of Sprague-Dawley rats obtained from two different suppliers, with the suppliers' rat colonies having drifted slightly apart in hemodynamic characteristics. Models used for the classification include deep neural network (DNN), random forest, support vector machine, multilayer perceptron. While the DNN uses raw pressure/flow measurements as features, the latter three use a feature vector of parameters of a nonlinear dynamic system fitted to the pressure/flow data, thereby restricting the classification basis to the hemodynamics. Although the performance in these cases is slightly reduced in comparison to that of the DNN, they still show promise for machine learning (ML) application. The pioneering contribution of this work is the establishment that even with features limited to hemodynamics-based information, the ML models can successfully achieve classification with reasonably high accuracy.

The L-NAME mouse model of preeclampsia and impact to long-term maternal cardiovascular health

Preeclampsia affects ∼2–8% of pregnancies worldwide. It is associated with increased long-term maternal cardiovascular disease risk. This study assesses the effect of the vasoconstrictor N(ω)-nitro-L-arginine methyl ester (L-NAME) in modelling preeclampsia in mice, and its long-term effects on maternal cardiovascular health. In this study, we found that L-NAME administration mimicked key characteristics of preeclampsia, including elevated blood pressure, impaired fetal and placental growth, and increased circulating endothelin-1 (vasoconstrictor), soluble fms-like tyrosine kinase-1 (anti-angiogenic factor), and C-reactive protein (inflammatory marker). Post-delivery, mice that received L-NAME in pregnancy recovered, with no discernible changes in measured cardiovascular indices at 1-, 2-, and 4-wk post-delivery, compared with matched controls. At 10-wk post-delivery, arteries collected from the L-NAME mice constricted significantly more to phenylephrine than controls. In addition, these mice had increased kidney Mmp9:Timp1 and heart Tnf mRNA expression, indicating increased inflammation. These findings suggest that though administration of L-NAME in mice certainly models key characteristics of preeclampsia during pregnancy, it does not appear to model the adverse increase in cardiovascular disease risk seen in individuals after preeclampsia.

Interactions between the complement and endothelin systems in normal pregnancy and following placental ischemia

Preeclampsia is characterized by new onset hypertension and fetal growth restriction and is associated with aberrant activation of the innate immune complement system and stressed or ischemic placenta. Previous studies have suggested a role for both endothelin and complement system activation products in new onset hypertension in pregnancy, but inter-relationships of the pathways are unclear. We hypothesized that complement activation following placental ischemia stimulates the endothelin pathway to cause hypertension and impair fetal growth. The Reduced Uterine Perfusion Pressure (RUPP) model results in hypertension and fetal growth restriction in a pregnant rat due to placental ischemia caused by mechanical obstruction of blood flow to uterus and placenta. The effect of inhibitor of complement activation soluble Complement Receptor 1 (sCR1) and endothelin A receptor (ET_A) antagonist atrasentan on hypertension, fetal weight, complement activation (systemic circulating C3a and local C3 placental deposition) and endothelin [circulating endothelin and message for preproendothelin (PPE), ET_A and endothelin B receptor (ET_B) in placenta] in the RUPP rat model were determined. Following placental ischemia, sCR1 attenuated hypertension but increased message for PPE and ET_A in placenta, suggesting complement activation causes hypertension via an endothelin independent pathway. With ET_A antagonism the placental ischemia-induced increase in circulating C3a was unaffected despite inhibition of hypertension, indicating systemic C3a alone is not sufficient. In normal pregnancy, inhibiting complement activation increased plasma endothelin but not placental PPE message. Atrasentan treatment increased fetal weight, circulating endothelin and placental ET_A message, and unexpectedly increased local complement activation in placenta (C3 deposition) but not C3a in circulation, suggesting endothelin controls local placental complement activation in normal pregnancy. Atrasentan also significantly decreased message for endogenous complement regulators Crry and CD55 in placenta and kidney in normal pregnancy. Results of our study indicate that complement/endothelin interactions differ in pregnancies complicated with placental ischemia vs normal pregnancy, as well as locally vs systemically. These data clearly illustrate the complex interplay between complement and endothelin indicating that perturbations of either pathway may affect pregnancy outcomes.

Neutrophil Depletion Attenuates Placental Ischemia-Induced Hypertension in the Rat

Preeclampsia is characterized by reduced placental perfusion with placental ischemia and hypertension during pregnancy. Preeclamptic women also exhibit a heightened inflammatory state and greater number of neutrophils in the vasculature compared to normal pregnancy. Since neutrophils are associated with tissue injury and inflammation, we hypothesized that neutrophils are critical to placental ischemia-induced hypertension and fetal demise. Using the reduced uteroplacental perfusion pressure (RUPP) model of placental ischemia-induced hypertension in the rat, we determined the effect of neutrophil depletion on blood pressure and fetal resorptions. Neutrophils were depleted with repeated injections of polyclonal rabbit anti-rat polymorphonuclear leukocyte (PMN) antibody (antiPMN). Rats received either antiPMN or normal rabbit serum (Control) on 13.5, 15.5, 17.5, and 18.5 days post conception (dpc). On 14.5 dpc, rats underwent either Sham surgery or clip placement on ovarian arteries and abdominal aorta to reduce uterine perfusion pressure (RUPP). On 18.5 dpc, carotid arterial catheters were placed and mean arterial pressure (MAP) was measured on 19.5 dpc. Neutrophil-depleted rats had reduced circulating neutrophils from 14.5 to 19.5 dpc compared to Control, as well as decreased neutrophils in lung and placenta on 19.5 dpc. MAP increased in RUPP Control vs Sham Control rats, and neutrophil depletion attenuated this increase in MAP in RUPP rats without any effect on Sham rats. The RUPP-induced increase in fetal resorptions and complement activation product C3a were not affected by neutrophil depletion. Thus, these data are the first to indicate that neutrophils play an important role in RUPP hypertension and that cells of the innate immune system may significantly contribute to pregnancy-induced hypertension.

Variation in the form of Pavlovian conditioned approach behavior among outbred male Sprague-Dawley rats from different vendors and colonies: sign-tracking vs. goal-tracking

Even when trained under exactly the same conditions outbred male Sprague-Dawley (SD) rats vary in the form of the Pavlovian conditioned approach response (CR) they acquire. The form of the CR (i.e. sign-tracking vs. goal-tracking) predicts to what degree individuals attribute incentive salience to cues associated with food or drugs. However, we have noticed variation in the incidence of these two phenotypes in rats obtained from different vendors. In this study, we quantified sign- and goal-tracking behavior in a reasonably large sample of SD rats obtained from two vendors (Harlan or Charles River), as well as from individual colonies operated by both vendors. Our sample of rats acquired from Harlan had, on average, more sign-trackers than goal-trackers, and vice versa for our sample of rats acquired from Charles River. Furthermore, there were significant differences among colonies of the same vendor. Although it is impossible to rule out environmental variables, SD rats at different vendors and barriers may have reduced phenotypic heterogeneity as a result of genetic variables, such as random genetic drift or population bottlenecks. Consistent with this hypothesis, we identified marked population structure among colonies from Harlan. Therefore, despite sharing the same name, investigators should be aware that important genetic and phenotypic differences exist among SD rats from different vendors or even from different colonies of the same vendor. If used judiciously this can be an asset to experimental design, but it can also be a pitfall for those unaware of the issue.

An acute rat in vivo screening model to predict compounds that alter blood glucose and/or insulin regulation

INTRODUCTION

Drug-induced glucose dysregulation and insulin resistance have been associated with weight gain and potential induction and/or exacerbation of diabetes mellitus in the clinic suggesting they may be safety biomarkers when developing antipsychotics. Glucose and insulin have also been suggested as potential efficacy biomarkers for some oncology compounds. The objective of this study was to qualify a medium throughput rat in vivo acute Intravenous Glucose Tolerance Test (IVGTT) for predicting compounds that will induce altered blood glucose and/or insulin levels.

METHODS

Acute and sub-chronic studies were performed to qualify an acute IVGTT model. Double cannulated male rats (Han-Wistar and Sprague-Dawley) were administered vehicle, olanzapine, aripiprazole or other compounds at t=-44min for acute studies and at time=-44min on the last day of dosing for sub-chronic studies, treated with dextrose (time=0min; i.v.) and blood collected using an automated Culex® system for glucose and insulin analysis (time=-45, -1, 2, 10, 15, 30, 45, 60, 75, 90, 120, 150 and 180min).

RESULTS

Olanzapine significantly increased glucose and insulin area under the curve (AUC) values while aripiprazole AUC values were similar to control, in both acute and sub-chronic studies. All atypical antipsychotics evaluated were consistent with literature references of clinical weight gain. As efficacy biomarkers, insulin AUC but not glucose AUC values were increased with a compound known to have insulin growth factor-1 (IGF-1) activity, compared to control treatment.

DISCUSSION

These studies qualified the medium throughput acute IVGTT model to more quickly screen compounds for 1) safety - the potential to elicit glucose dysregulation and/or insulin resistance and 2) efficacy - as a surrogate for compounds affecting the glucose and/or insulin regulatory pathways. These data demonstrate that the same in vivo rat model and assays can be used to predict both clinical safety and efficacy of compounds.

Complement activation is critical for placental ischemia-induced hypertension in the rat

Preeclampsia is a major obstetric problem defined by new-onset hypertension and proteinuria associated with compromised placental perfusion. Although activation of the complement system is increased in preeclampsia compared to normal pregnancy, it remains unclear whether excess complement activation is a cause or consequence of placental ischemia. Therefore, we hypothesized that complement activation is critical for placental ischemia-induced hypertension. We employed the reduced utero-placental perfusion pressure (RUPP) model of placental ischemia in the rat to induce hypertension in the third trimester and evaluated the effect of inhibiting complement activation with a soluble recombinant form of an endogenous complement regulator, human complement receptor 1 (sCR1; CDX-1135). On day 14 of a 21-day gestation, rats received either RUPP or Sham surgery and 15 mg/kg/day sCR1 or saline intravenously on days 14-18. Circulating complement component 3 decreased and complement activation product C3a increased in RUPP vs. Sham (p<0.05), indicating complement activation had occurred. Mean arterial pressure (MAP) measured on day 19 increased in RUPP vs. Sham rats (109.8±2.8 mmHg vs. 93.6±1.6 mmHg). Treatment with sCR1 significantly reduced elevated MAP in RUPP rats (98.4±3.6 mmHg, p<0.05) and reduced C3a production. Vascular endothelial growth factor (VEGF) decreased in RUPP compared to Sham rats, and the decrease in VEGF was not affected by sCR1 treatment. Thus, these studies have identified a mechanistic link between complement activation and the pregnancy complication of hypertension apart from free plasma VEGF and have identified complement inhibition as a potential treatment strategy for placental ischemia-induced hypertension in preeclampsia.

Large BP-dependent and -independent differences in susceptibility to nephropathy after nitric oxide inhibition in Sprague-Dawley rats from two major suppliers

The N(ω)-nitro-l-arginine methyl ester (l-NAME) model is widely employed to investigate the role of nitric oxide (NO) in renal injury. The present studies show that Sprague-Dawley rats from Harlan (H) and Charles River (CR) exhibit strikingly large differences in susceptibility to l-NAME nephropathy. After 4 wk of l-NAME (∼50 mg·kg(-1)·day(-1) in drinking water), H rats (n = 13) exhibited the expected hypertension [average radiotelemetric systolic blood pressure (BP), 180 ± 3 mmHg], proteinuria (136 ± 17 mg/24 h), and glomerular injury (GI) (12 ± 2%). By contrast, CR rats developed less hypertension (142 ± 4), but surprisingly no proteinuria or GI, indicating a lack of glomerular hypertension. Additional studies showed that conscious H, but not CR, rats exhibit dose-dependent renal vasoconstriction after l-NAME. To further investigate these susceptibility differences, l-NAME was given 2 wk after 3/4 normotensive nephrectomy (NX) and comparably impaired renal autoregulation in CR-NX and H-NX rats. CR-NX rats, nevertheless, still failed to develop proteinuria and GI despite moderate hypertension (144 ± 2 mmHg, n = 29). By contrast, despite an 80-90% l-NAME dose reduction and lesser BP increases (169 ± 4 mmHg), H-NX rats (n = 20) developed greater GI (26 ± 3%) compared with intact H rats. Linear regression analysis showed significant (P < 0.01) differences in the slope of the relationship between BP and GI between H-NX (slope 0.56 ± 0.14; r = 0.69; P < 0.008) and CR-NX (slope 0.09 ± 0.06; r = 0.29; P = 0.12) rats. These data indicate that blunted BP responses to l-NAME in the CR rats are associated with BP-independent resistance to nephropathy, possibly mediated by a resistance to the renal (efferent arteriolar) vasoconstrictive effects of NO inhibition.

Determining normal variability in a developmental neurotoxicity test: a report from the ILSI Research Foundation/Risk Science Institute expert working group on neurodevelopmental endpoints

With the implementation of the Food Quality Protection Act in 1996, more detailed evaluations of possible health effects of pesticides on developing organisms have been required. As a result, considerable developmental neurotoxicity (DNT) data have been generated on a variety of endpoints, including developmental changes in motor activity, auditory startle habituation, and various learning and memory parameters. One issue in interpreting these data is the level of variability for the measures used in these studies: excessive variability can obscure treatment-related effects, or conversely, small but statistically significant changes could be viewed as treatment related, when they might in fact be within the normal range. To aid laboratories in designing useful DNT studies for regulatory consideration, an operational framework for evaluating observed variability in study data has been developed. Elements of the framework suggest how an investigator might approach characterization of variability in the dataset; identification of appropriate datasets for comparison; evaluation of similarities and differences in variability between these datasets, and of possible sources of the variability, including those related to test conduct and test design. A case study using auditory startle habituation data is then presented, employing the elements of this proposed approach.

Fetal nucleated red blood cells in a rat model of intrauterine growth restriction induced by hypoxia and nitric oxide synthase inhibition

OBJECTIVE

Nucleated red blood cells (NRBCs) in fetal circulation have been proposed as a marker of chronic hypoxia in fetuses with intrauterine growth restriction (IUGR). We sought to determine the effects of chronic hypoxia, chronic nitric oxide inhibition with N(G)-nitro-L-arginine methyl ester (L-NAME), or both on NRBC counts, erythropoietin levels, and pathologic changes in an animal model of IUGR.

STUDY DESIGN

We assigned timed pregnant adult Sprague Dawley rats to the following groups: (1) 21% oxygen + saline solution (n = 7); (2) 21% oxygen + L-NAME (n = 8); (3) 10% oxygen + saline solution (n = 6); and (4) 10% oxygen + L-NAME (n = 6). We inserted osmotic pumps that were prefilled with saline solution or L-NAME subcutaneously on day 17 of gestation. The animals were placed in a Plexiglas hypoxic chamber, which ensured a constant hypoxic environment. The animals were killed on day 21 of gestation before the onset of spontaneous labor. We collected maternal and fetal blood for measurement of NRBC and erythropoietin levels. The results were interpreted in relationship to maternal arterial blood gases and hemoglobin and hematocrit levels. Fetuses were examined for gross abnormalities and histological abnormalities that are characteristic of vascular disruptions by a blind examiner to experimental manipulation.

RESULTS

Nitric oxide inhibition induced IUGR with maximal effect when both L-NAME and hypoxia treatments were combined. Inhibition of nitric oxide synthesis, but not chronic hypoxia, increased the number of fetal NRBCs and generalized hemorrhagic diathesis in utero. These features were aggravated significantly when the treatments were combined. Moreover, chronic hypoxia induced significant maternal metabolic acidosis and increased hematocrit and erythropoietin levels in maternal and fetal blood. Nitric oxide inhibition increased maternal hematocrit levels while decreasing maternal erythropoietin levels without significantly altering the maternal acid-base status. In contrast with chronic hypoxia, nitric oxide inhibition increased fetal NRBCs without affecting erythropoietin levels.

CONCLUSION

Our findings indicate that the number of NRBCs in fetal circulation does not serve as a specific marker of chronic hypoxia that accompanies IUGR or of elevated erythropoietin levels but are an epiphenomenon that is related to the inhibition of nitric oxide.

Placentomal differentiation may compensate for maternal nutrient restriction in ewes adapted to harsh range conditions

Maternal nutrient restriction from early to midgestation can lead to fetal growth retardation, with long-term impacts on offspring growth, physiology, and metabolism. We hypothesized that ewes from flocks managed under markedly different environmental conditions and levels of nutrition might differ in their ability to protect their own fetus from a bout of maternal nutrient restriction. We utilized multiparous ewes of similar breeding, age, and parity from 2 flocks managed as 1) ewes adapted to a nomadic existence and year-long, limited nutrition near Baggs, WY (Baggs ewes), and 2) University of Wyoming ewes with a sedentary lifestyle and continuous provision of more than adequate nutrition (UW ewes). Groups of Baggs ewes and UW ewes were fed 50 (nutrient restricted) or 100% (control fed) of National Research Council recommendations from d 28 to 78 of gestation, then necropsied, and fetal and placental data were obtained. Although there was a marked decrease (P < 0.05) in fetal weight and blood glucose concentrations in nutrient-restricted vs. control fed UW ewes, there was no difference in these fetal measurements between nutrient-restricted and control-fed Baggs ewes. Nutrient-restricted and control-fed UW ewes exhibited predominantly type A placentomes on d 78, but there were fewer (P c0.05) type A and greater (P < 0.05) numbers of type B, C, and D placentomes in nutrient-restricted than control-fed Baggs ewes. Placental efficiency (fetal weight/placentomal weight) was reduced (P = 0.04) in d 78 nutrient-restricted UW ewes when compared with control-fed UW ewes. In contrast, nutrient-restricted and control-fed Baggs ewes exhibited similar placental efficiencies on d 78. This is the first report of different placental responses to a nutritional challenge during pregnancy when ewes were selected under different management systems. These data are consistent with the concept that Baggs ewes or their conceptuses, which were adapted to both harsh environments and limited nutrition, initiated conversion of type A placentomes to other placentomal types when subjected to an early to mid-gestational nutrient restriction, whereas this conversion failed to occur in UW ewes. This early placentomal conversion in the Baggs ewes may function to maintain normal nutrient delivery to their developing fetuses during maternal nutrient restriction.

Magnesium sulfate effectively reduces blood pressure in an animal model of preeclampsia

OBJECTIVE

We tested the ability of magnesium sulfate to reduce hypertension and neonatal growth retardation in an animal model of preeclampsia.

STUDY DESIGN

On day 17 of pregnancy, osmotic minipumps were inserted subcutaneously to continuously deliver either vehicle (saline control group), or N-nitro-L-arginine methyl ester (L-NAME) (50 mg/kg/day), or L-NAME (50 mg/kg/day) in combination with magnesium sulfate (60 mg/kg/day). Prior to insertion, blood pressure and heart rate were monitored with a pneumatic tail cuff device. Blood pressure measurements were repeated on days 18, 20, and 21 of pregnancy. Blood was obtained on days 17 and 21, along with urine, to assess magnesium levels and degree of proteinuria. Pups were weighed and measured at 48 hours postpartum.

RESULTS

Rats receiving L-NAME developed hypertension within 24 hours of implantation (108 +/- 3.9 vs. 123 +/- 3.4 mmHg, p < 0.05). Magnesium sulfate, given along with L-NAME did not prevent mean blood pressure from increasing, but reduced it by day 21 compared to L-NAME given alone (107 +/- 3.4 vs. 122 +/- 8.7 mmHg, respectively, p < 0.05). Magnesium sulfate reduced neonatal growth retardation by improving the weight of the pups compared to pups from maternal rats given L-NAME alone (6.1 +/- 0.1 vs. 5.2 +/- 0.3 grams, respectively, p < 0.05).

CONCLUSION

Maternal magnesium sulfate reduces blood pressure and increases neonatal size compared to L-NAME without magnesium. These findings support a beneficial effect of magnesium in preeclampsia.

MgSO4 prevents left ventricular dysfunction in an animal model of preeclampsia

OBJECTIVE

We hypothesized that cardiac function would be reduced in a pregnant rat model of preeclampsia induced by L-NAME, a NOS inhibitor, and be reversed with magnesium sulfate prophylaxis.

STUDY DESIGN

Female Sprague-Dawley rats were bred in-house. On gestational day 17, rats were anesthetized and osmotic minipumps were implanted to continuously deliver saline, L-NAME, or L-NAME and MgSO4. On gestational day 21, hearts were isolated and perfused in the working mode using Krebs Henseleit buffer.

RESULTS

Pregnant rats treated with L-NAME displayed significant hypertension compared to the saline-treated controls (P < 0.05). Moreover, cardiac output and cardiac work were significantly reduced in the L-NAME-treated rats compared to controls (P < 0.05). In the L-NAME-treated rats given MgSO4, cardiac function remained normal.

CONCLUSION

Cardiac function is depressed in an animal model of preeclampsia induced by L-NAME infusion. MgSO4 prevented the reduction in cardiac function and is clearly beneficial in preserving normal heart function in preeclampsia.

Placental dysfunction in Suramin-treated rats – a new model for pre-eclampsia

Impaired placentation and oxidative stress are proposed to play major roles in the pathogenesis of placental dysfunction and pre-eclampsia. This study was carried out to evaluate if inhibited angiogenesis by Suramin injections in early pregnancy may cause a condition resembling pre-eclampsia in rats. Rats of two different Sprague-Dawley strains, U and H, were given intraperitoneal injections of Suramin or saline in early pregnancy. The outcome of pregnancy was evaluated on gestational day 20. Suramin injections caused increased blood pressure and decreased renal blood flow in the U rats. In both rat strains Suramin decreased the placental blood flow and caused fetal growth retardation. In both strains the placental concentration of the isoprostane 8-epi-PGF2alpha was increased, indicating oxidative stress. The serum concentration of Endothelin-1 was increased in the U rats. The U strain had a lower basal placental blood flow, and the effects of Suramin were more pronounced in this strain. We conclude, that Suramin injections to pregnant rats cause a state of placental insufficiency, which partly resembles human pre-eclampsia. The induction of this condition is at least partly mediated by oxidative stress, and is subject to varied genetic susceptibility.

Animal models of obstetric complications in relation to schizophrenia

Epidemiological studies have provided strong evidence that exposure to obstetric complications is associated with an increased risk for later development of schizophrenia. These human studies have now begun to tease out which specific pregnancy, labor/delivery or neonatal complications might confer greatest risk for schizophrenia. Animal modeling can be a useful tool to directly ask if a particular obstetric complication can actually cause changes in brain function or behavior resembling changes in schizophrenia. This review describes currently available animal models for some of the obstetric complications with greatest effect size for schizophrenia, including maternal diabetes, preeclampsia, infection and stress during pregnancy, intrauterine growth retardation and fetal/neonatal hypoxia. Where available, evidence that these types of obstetric complications in animals produce alterations in CNS function or behavior, related to features of schizophrenic pathology, is presented. Animal models might provide insights into the mechanisms by which specific obstetric complications have long-term influence on brain development leading to increased risk for schizophrenia. Factors common to several obstetric complications associated with schizophrenia may also be discerned. In this way, animal modeling may provide the framework for human studies to ask further more refined questions concerning the role of specific obstetric factors contributing to schizophrenia, and may provide clues to prevention.

Incidence and severity of neovascularization in oxygen- and metabolic acidosis-induced retinopathy depend on rat source

PURPOSE

We have previously described oxygen-induced retinopathy (OIR) and metabolic acidosis-induced retinopathy (MAIR) in the neonatal rat, both of which are analogous to human retinopathy of prematurity (ROP). Given that rats of identical strain from two commercial suppliers are phenotypically different, we investigated the incidence and severity of preretinal neovascularization (NV) in rats from different suppliers using the OIR and MAIR models.

METHODS

Using our established models for OIR and MAIR, 400 newborn Sprague-Dawley rats, obtained from Harlan Laboratories (HSD) and Charles River Laboratories (CRSD), were raised in 16 expanded litters of 25 (6 OIR and 10 MAIR). Beginning at day 1 of life, OIR litters (3 HSD, 3 CRSD) were exposed to 7 daily cycles of hyperoxia (80% O(2), 20.5 hours) and hypoxia (10% O(2), 0.5 hours) with a gradual return to 80% O(2) over 3 hours. OIR rats were sacrificed after 5 days of room air recovery. MAIR litters (5 HSD, 5 CRSD) were raised in room air and gavaged twice daily with NH(4)Cl (10 mM/kg body weight) from day 2 through day 4. MAIR rats were sacrificed after 3 days of recovery. For both OIR and MAIR litters, retinae from left eyes were dissected, ADPase-stained, and flatmounted. Presence and severity of NV was scored and retinal vascular areas measured by a masked observer.

RESULTS

In OIR rats, the incidence of NV was higher in CRSD rats than HSD rats (73% vs. 45%, p = 0.002). NV was more severe in CRSD rats than HSD rats (median clock hours 2 vs. 0, p = 0.0001). In MAIR rats, the incidence of NV was comparable between CRSD and HSD rats (29% vs. 34%, p = 0.53) and there was no significant difference in the severity of NV.

CONCLUSIONS

Sprague-Dawley rats obtained from two independent commercial sources differed in their incidence and severity of NV associated with OIR, but not with MAIR. Future genetic studies are warranted to investigate the differences between CRSD and HSD rats, which might yield further clues into the pathogenesis of ROP.