Developing a Qualification and Verification Strategy for Digital Tissue Image Analysis in Toxicological Pathology.

Digital tissue image analysis is a computational method for analyzing whole-slide images and extracting large, complex, and quantitative data sets. However, as with any analysis method, the quality of generated results is dependent on a well-designed quality control system for the entire digital pathology workflow. Such system requires clear procedural controls, appropriate user training, and involvement of specialists to oversee key steps of the workflow. The toxicologic pathologist is responsible for reporting data obtained by digital image analysis and therefore needs to ensure that it is correct. To accomplish that, they must understand the main parameters of the quality control system and should play an integral part in its conception and implementation. This manuscript describes the most common digital tissue image analysis end points and potential sources of analysis errors. In addition, it outlines recommended approaches for ensuring quality and correctness of results for both classical and machine-learning based image analysis solutions, as adapted from a recently proposed Food and Drug Administration regulatory framework for modifications to artificial intelligence/machine learning-based software as a medical device. These approaches are beneficial for any type of toxicopathologic study which uses the described end points and can be adjusted based on the intended use of the image analysis solution.

Differential Receptor Tyrosine Kinase Phosphorylation in the Uterus of Rats Following Developmental Exposure to Tetrabromobisphenol A.

Tetrabromobisphenol A (TBBPA) is a brominated flame retardant that induces endometrial adenocarcinoma and other uterine tumors in Wistar Han rats; however, early molecular events or biomarkers of TBBPA exposure remain unknown. We investigated the effects of TBBPA on growth factor receptor activation (phospho-RTK) in uteri of rats following early-life exposures. Pregnant Wistar Han rats were exposed to TBBPA (0, 0.1, 25, 250 mg/kg/day) via oral gavage on gestation day 6 through weaning of pups (PND 21). Pups were exposed in utero, through lactation, and by daily gavage from PND 22 to PND 90. Uterine horns were collected (at PND 21, PND 33, PND 90) and formalin-fixed or frozen for histologic, immunohistochemical, phospho-RTK arrays, or western blot analysis. At PND 21, the phosphor-RTKs, FGFR2, FGFR3, TRKC and EPHA1 were significantly increased at different treatment concentrations. Several phospho-RTKs were also significantly overexpressed at PND 33 which included epithelial growth factor receptor (EGFR), Fibroblast Growth Factor Receptor 3-4 (FGFR2, FGFR3, FGFR4), insulin-like growth factor receptor 1 (IGF1R), INSR, AXL, MERTK, PDGFRa and b, RET, Tyrosine Kinase with Immunoglobulin Like and EGF Like Domains 1 and 2 (TIE1; TIE2), TRKA, VEGFR2 and 3, and EPHA1 at different dose treatments. EGFR, an RTK overexpressed in endometrial cancer in women, remained significantly increased for all treatment groups at PND 90. Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2) and IGF1R were overexpressed at PND 33 and remained increased through PND 90, although ERBB2 was statistically significant at PND 90. The phospho-RTKs, FGFR3, AXL, DTK, HGFR, TRKC, VEGFR1 and EPHB2 and 4 were also statistically significant at PND 90 at different dose treatments. The downstream effector, phospho-MAPK44/42 was also increased in uteri of treated rats. Our findings show RTKs are dysregulated following early life TBBPA exposures and their sustained activation may contribute to TBBPA-induced uterine tumors observed in rats later in life.

Qualitative and Quantitative Neuropathology Approaches Using Magnetic Resonance Microscopy (Diffusion Tensor Imaging) and Stereology in a Hexachlorophene Model of Myelinopathy in Sprague-Dawley Rats.

It is well established that hexachlorophene, which is used as an antibacterial agent, causes intramyelinic edema in humans and animal models. The hexachlorophene myelinopathy model, in which male Sprague-Dawley rats received 25 to 30 mg/kg hexachlorophene by gavage for up to 5 days, provided an opportunity to compare traditional neuropathology evaluations with magnetic resonance microscopy (MRM) findings. In addition, stereology assessments of 3 neuroanatomical sites were compared to quantitative measurements of similar structures by MRM. There were positive correlations between hematoxylin and eosin and luxol fast blue stains and MRM for identifying intramyelinic edema in the cingulum of corpus callosum, optic chiasm, anterior commissure (aca), lateral olfactory tracts, pyramidal tracts (py), and white matter tracts in the cerebellum. Stereology assessments were focused on the aca, longitudinal fasciculus of the pons, and py and demonstrated differences between control and treated rats, as was observed using MRM. The added value of MRM assessments was the ability to acquire qualitative 3-dimensional (3-D) images and obtain quantitative measurements of intramyelinic edema in 26 neuroanatomical sites in the intact brain. Also, diffusion tensor imaging (fractional anisotropy [FA]) indicated that there were changes in the cytoarchitecture of the white matter as detected by decreases in the FA in the treated compared to the control rats. This study demonstrates creative strategies that are possible using qualitative and quantitative assessments of potential white matter neurotoxicants in nonclinical toxicity studies. Our results lead us to the conclusion that volumetric analysis by MRM and stereology adds significant value to the standard 2-D microscopic evaluations.

Stereology of the Peripheral Nervous System.

Qualitative histopathology has been the gold standard for evaluation of morphological tissue changes in all organ systems, including the peripheral nervous system. However, the human eye is not sensitive enough to detect small changes in quantity or size. Peripheral nervous system toxicity can manifest as subtle changes in neuron size, neuron number, axon size, number of myelinated or unmyelinated axons, or number of nerve fibers. Detection of these changes may be beyond the sensitivity of the human eye alone, necessitating quantitative approaches in some cases. Although 2-dimensional (2D) histomorphometry can provide additional information and is more sensitive than qualitative evaluation alone, the results are not always representative of the entire tissue and assumptions about the tissue can lead to bias, or inaccuracies, in the data. Design-based stereology provides 3D estimates of number, volume, surface area, or length, and stereological principles can be applied to peripheral nervous system tissues to obtain accurate and precise estimates, such as neuron number and size, axon number, and total intraepidermal nerve fiber length. This review describes practical stereological approaches to 3 compartments of the peripheral nervous system: ganglia, peripheral nerves, and intraepidermal nerve fibers.

Phenotyping of adipose, liver, and skeletal muscle insulin resistance and response to pioglitazone in spontaneously obese rhesus monkeys.

Insulin resistance and diabetes can develop spontaneously with obesity and aging in rhesus monkeys, highly similar to the natural history of obesity, insulin resistance, and progression to type 2 diabetes in humans. The current studies in obese rhesus were undertaken to assess hepatic and adipose contributions to systemic insulin resistance-currently, a gap in our knowledge-and to benchmark the responses to pioglitazone (PIO). A two-step hyperinsulinemic-euglycemic clamp, with tracer-based glucose flux estimates, was used to measure insulin resistance, and in an intervention study was repeated following 6 wk of PIO treatment (3 mg/kg). Compared with lean healthy rhesus, obese rhesus has a 60% reduction of glucose utilization during a high insulin infusion and markedly impaired suppression of lipolysis, which was evident at both low and high insulin infusion. However, obese dysmetabolic rhesus manifests only mild hepatic insulin resistance. Six-week PIO treatment significantly improved skeletal muscle and adipose insulin resistance (by ~50%). These studies strengthen the concept that insulin resistance in obese rhesus closely resembles human insulin resistance and indicate the value of obese rhesus for appraising new insulin-sensitizing therapeutics.

Characterization of Metabolic Status in Nonhuman Primates with the Intravenous Glucose Tolerance Test.

The intravenous glucose tolerance test (IVGTT) plays a key role in the characterization of glucose homeostasis. When taken together with serum biochemical profiles, inclusive of blood glucose levels in both the fed and fasted state, HbA1c, insulin levels, clinical history of diet, body composition, and body weight status, an assessment of normal and abnormal glycemic control can be made. Interpretation of an IVGTT is done through measurement of changes in glucose and insulin levels over time in relation to the dextrose challenge. Critical components to be considered are: peak glucose and insulin levels reached in relation to T0 (end of glucose infusion), the glucose clearance rate K derived from the slope of rapid glucose clearance in the first 20 min (T1 to T20), the time to return to glucose baseline, and the area under the curve (AUC). These IVGTT measures will show characteristic changes as glucose homeostasis moves from a healthy to a diseased metabolic state5. Herein we will describe the characterization of nonhuman primates (Rhesus and Cynomolgus macaques), which are the most relevant animal model of Type II diabetes (T2D) in humans and the IVGTT and clinical profiles of these animals from a lean healthy, to obese dysmetabolic, and T2D state 8, 10, 11.

Left ventricular diastolic dysfunction in nonhuman primate model of dysmetabolism and diabetes.

BACKGROUND: Diabetes is one of the major risk factors for cardiomyopathy and heart failure with reduced ejection fraction (EF) and highly associated with left ventricular (LV) dysfunction in human. This study aimed 1) to noninvasively assess cardiac function using echocardiography; 2) to test the hypothesis that like diabetic human, cardiac function may also be compromised; in spontaneously developed obese, dysmetabolic and diabetic nonhuman primates (NHPs). METHODS: Cardiovascular functions were measured by noninvasive echocardiography in 28 control, 20 dysmetabolic/pre-diabetic and 41 diabetic cynomolgus monkeys based on fasting blood glucose and other metabolic status. RESULTS: The LV end-systolic volume (ESV) was higher while end-diastolic volume (EDV, 12 ± 5.7 mL) and EF (63 ± 12.8 %) significantly lower in the diabetic compared to control (14 ± 7 mL and 68 ± 9.8 %) group, respectively. The E/A ratio of LV trans-mitral peak flow rate during early (E) over late (A) diastole was significantly lower in the diabetic (1.19 ± 0.45) than control (1.44 ± 0.48) group. E-wave deceleration time (E DT) was prolonged in the diabetic (89 ± 41 ms) compared to control (78 ± 26 ms) group. Left atrial (LA) maximal dimension (LADmax) was significantly greater in the diabetic (1.3 ± 0.17 cm) than control (1.1 ± 0.16 cm) group. Biochemical tests showed that total cholesterol and LDL were significant higher in the diabetic (167 ± 63 and 69 ± 37 mg/dL) than both pre-diabetic (113 ± 37 and 41 ± 23 mg/dL) and control (120 ± 28 and 41 ± 17 mg/dL) groups, respectively. Multivariable logistic regression analysis demonstrated that LV systolic (reduced EF) and diastolic (abnormal E/A ratio) dysfunctions are significantly correlated with aging and hyperglycemia. Histopathology examination of the necropsy heart revealed inflammatory infiltration, cardiomyocyte hypertrophy and fragmentation, indicating the myocardial ischemia and remodeling which is consistent with the LV dysfunction phenotype. CONCLUSIONS: Using noninvasive echocardiography, the present study demonstrated for the first time that dysmetabolic and diabetic NHPs are associated with LV systolic (increased ESV, decreased EF, etc.) and diastolic (decreased EDV and E/A ratio, prolonged E DT, etc.) dysfunctions, accompanied by LA hypertrophic remodeling (increased LADmax), the phenotypes similarly to those found in diabetic patients. Thus, spontaneously developed dysmetabolic and diabetic NHPs is a highly translatable model to human diseases not only in the pathogenic mechanisms but also can be used for testing novel therapies for cardiometabolic disorders.

Corticotropin-releasing factor receptor-dependent effects of repeated stress on tau phosphorylation, solubility, and aggregation.

Exposure and/or sensitivity to stress have been implicated as conferring risk for development of Alzheimer's disease (AD). Although the basis for such a link remains unclear, we previously reported differential involvement of corticotropin-releasing factor receptor (CRFR) 1 and 2 in acute stress-induced tau phosphorylation (tau-P) and solubility in the hippocampus. Here we examined the role of CRFRs in tau-P induced by repeated stress and the structural manifestations of altered tau solubility. Robust tau-P responses were seen in WT and CRFR2 null mice exposed to repeated stress, which were sustained at even 24 h after the final stress exposure. A portion of phosphorylated tau in these mice was sequestered in detergent-soluble cellular fractions. In contrast, CRFR1 and CRFR double-KO mice did not exhibit repeated stress-induced alterations in tau-P or solubility. Similarly, treatment with CRFR1 antagonist attenuated repeated stress-induced tau-P. Using histochemical approaches in a transgenic CRFR1 reporter mouse line, we found substantial overlap between hippocampal CRFR1 expression and cells positive for phosphorylated tau after exposure to repeated stress. Ultrastructural analysis of negatively stained extracts from WT and CRFR2 null mice identified globular aggregates that displayed positive immunogold labeling for tau-P, as well as conformational changes in tau (MC1) seen in early AD. Given that repeated stress exposure results in chronic increases in hippocampal tau-P and its sequestration in an insoluble (and potentially prepathogenic) form, our data may define a link between stress and an AD-related pathogenic mechanism.

Lipopolysaccharide-induced tau phosphorylation and kinase activity--modulation, but not mediation, by corticotropin-releasing factor receptors.

Clinical studies suggest that exposure to stress can increase risk for Alzheimer's disease (AD). Although the precise links between stress and vulnerability to develop AD remain uncertain, recent animal work suggests that stress may promote susceptibility to AD pathology by activating tau kinases and inducing tau phosphorylation (tau-P). Our previous findings indicate the differential involvement of corticotropin-releasing factor receptor (CRFR) types 1 and 2 in regulating tau-P in the hippocampus induced by acute restraint, an emotional stressor. To assess the generality of CRFR involvement in stress-induced tau-P and tau kinase activity, the present study extends our investigation to a well-characterized physiological stressor, i.e. immune challenge induced by bacterial lipopolysaccharide (LPS). Acute systemic administration of LPS (100 µg/kg) robustly increased hippocampal (but not isocortical or cerebellar) tau-P, peaking at 40-120 min postinjection and abating thereafter. Assessments of the genotype dependence of this effect yielded results that were distinct from the restraint model. Treatment with LPS increased phosphorylation in wild-type, single and double CRFR knockouts with only subtle variation, which included a reliable exaggeration of tau-P responses in CRFR1-deficient mice. Parallel analyses implicated glycogen synthase kinase-3 and cyclin-dependent kinase-5 as likely cellular mediators of LPS-induced tau-P. Conversely, our data suggest that temperature-dependent fluctuations in tau protein phosphatase 2A (PP2A) may not play a role in this context. Thus, neither the strict CRFR1 dependence of restraint-induced tau-P nor the exaggeration of these responses in CRFR2 null mice generalize to the LPS model. CRFR mediation of stress-induced hippocampal tau-P may be limited to emotional stressors.

Ontogeny of L-DOPA-induced locomotion: Expression of c-Fos in the brainstem and basal ganglia of rats.

Neonatal rats suspended in harnesses, limbs hanging freely, and injected with 100 mg/kg l-3,4-dihydroxyphenylalanine (L-DOPA), engage in a behavior (air stepping) that closely resembles spontaneous locomotion. Rats no longer demonstrate this response after postnatal day 20 (P20). In the present experiment, an immunohistochemical analysis of the immediate early protein c-Fos was performed as a marker for cellular activity in the brains of suspended rat pups treated with l-DOPA at P15 and P25. Control rats were injected with saline at each age and subjected to the same behavioral protocol. Only P15 rat pups injected with L-DOPA engaged in air stepping and expressed the highest levels of c-Fos reactivity in output nuclei of the basal ganglia, as well as the pedunculopontine (PPN) and cuneiform (Cnf) nuclei. Twenty-five-day-old rats, which did not air step, exhibited reduced c-Fos labeling in these areas as well as in the locus coeruleus (LC). Our findings suggest that excitation of the basal ganglia resulted via afferents from the PPN and/or Cnf, which may develop before reciprocal inhibitory connections are fully mature. We propose that a circumscribed portion of the midbrain, which overlaps with the physiologically defined mesencephalic locomotor region (MLR), is necessary for the production of L-DOPA-induced locomotion. We propose further that this action is induced against a background of heightened arousal during the first three postnatal weeks but comes under inhibitory control in rat pups older than 20 days of age.