A novel workflow for unbiased 3D quantification of autophagosomes in Arabidopsis thaliana roots.

Macroautophagy is often quantified by live imaging of autophagosomes labeled with fluorescently tagged ATG8 protein (FP-ATG8) in Arabidopsis thaliana. The labeled particles are then counted in single focal planes. This approach may lead to inaccurate results as the actual 3D distribution of autophagosomes is not taken into account and appropriate sampling in the Z-direction is not performed. To overcome this issue, we developed a workflow consisting of immunolabeling of autophagosomes with an anti-ATG8 antibody followed by stereological image analysis using the optical disector and the Cavalieri principle. Our protocol specifically recognized autophagosomes in epidermal cells of Arabidopsis root. Since the anti-ATG8 antibody recognizes multiple AtATG8 isoforms, we were able to detect a higher number of immunolabeled autophagosomes than with the FP-AtATG8e marker, that most probably does not recognize all autophagosomes in a cell. The number of autophagosomes per tissue volume positively correlated with the intensity of autophagy induction. Compared with the quantification of autophagosomes in maximum intensity projections, stereological methods were able to detect the autophagosomes present in a given volume with higher accuracy. Our novel workflow provides a powerful toolkit for unbiased and reproducible quantification of autophagosomes and offers a convenient alternative to the standard of live imaging with FP-ATG8 markers.

Histology and design-based estimation of hepatocellularity and volumes of hepatocytes in control and ethynylestradiol exposed males of platyfish (Xiphophorus maculatus).

The liver hosts numerous vital functions, such as biotransformation and excretion of xenobiotics. Synthetic oestrogens influence liver structure and function, leading to adaptations or to dysfunctions/injury. They are often stated to induce increases in fish liver weight, but there is controversy regarding how: if by changes in hepatocyte size (hypertrophy) and/or number (hyperplasia). Using platyfish as the experimental model, our primary aim was to assess if/how hepatocytes reacted to a sub-acute oestrogenic exposure. A complementary aim was to generate fundamental structural data for the liver of that model organism. Adult males were injected intramuscularly with 17α-ethinylestradiol (EE2) (25 µg/g), every 72 h for two weeks. Control fish were given solvent only. Body and liver morphometry were registered, and hepatocytes examined through histology and stereology at light microscopy. Immunohistochemistry evaluated hepatocytic vitellogenin (VTG) content. Treated and control fish did not differ as to quantitative parameters. Nevertheless, exposed fish were sensitive to EE2. VTG tagging was positive in their hepatocytes and these tended to be more basophilic, though not fully oestrogenized. We hypothesise that the platyfish liver is not particularly sensitive to the disrupting action of EE2 because of its reproductive mode; with no production peaks of VTG and no huge changes in endogenous sex-steroids. The fish may have had no evolutionary pressure for hepatocytes to be particularly reactive to oestradiol (E2). In the end, this study offers the first unbiased estimation of the liver cellularity in the platyfish, as well of the hepatocytic volume, serving now as a baseline reference.

Numerical density of oligodendrocytes and oligodendrocyte clusters in the anterior putamen in major psychiatric disorders.

There is increasing evidence to support the notion that oligodendrocyte and myelin abnormalities may contribute to the functional dysconnectivity found in the major psychiatric disorders. The putamen, which is an important hub in the cortico-striato-thalamo-cortical loop, has been implicated in a broad spectrum of psychiatric illnesses and is a central target of their treatments. Previously we reported a reduction in the numerical density of oligodendrocytes and oligodendrocyte clusters in the prefrontal and parietal cortex in schizophrenia. Oligodendrocyte clusters contain oligodendrocyte progenitors and are involved in functionally dependent myelination. We measured the numerical density (Nv) of oligodendrocytes and oligodendrocyte clusters in the putamen in schizophrenia, bipolar disorder (BPD) and major depressive disorder (MDD) as compared to healthy controls (15 cases per group). Optical disector was used to estimate the Nv of oligodendrocytes and oligodendrocyte clusters. A significant reduction in both the Nv of oligodendrocytes (- 34%; p < 0.01) and the Nv of oligodendrocyte clusters (- 41%; p < 0.05) was found in the schizophrenia group as compared to the control group. Sexual dimorphism for both measurements was found only within the control group. The Nv of oligodendrocytes was significantly lower in male schizophrenia cases as compared to the male control cases. However, the Nv of oligodendrocyte clusters was significantly lower in all male clinical cases as compared to the male control group. The data suggest that lowered density of oligodendrocytes and oligodendrocyte clusters may contribute to the altered functional connectivity in the putamen in subjects with schizophrenia.

Reduced density of oligodendrocytes and oligodendrocyte clusters in the caudate nucleus in major psychiatric illnesses.

Functional dysconnectivity in schizophrenia and affective disorders may be associated with myelin and oligodendrocyte abnormalities. Altered network integration involving the caudate nucleus (CN) and metabolic abnormalities in fronto-striatal-thalamic white matter tracts have been reported in schizophrenia and impaired patterns of cortico-caudate functional connectivity have been found in both bipolar disorder (BPD) and schizophrenia compared to healthy controls. Postmortem studies have found ultrastructural dystrophy and degeneration of oligodendrocytes and dysmyelination in the CN in schizophrenia and BPD. We aimed to test the hypothesis that oligodendrocyte density may be reduced in the CN in major psychiatric disorders and may thereby form the cellular basis for the functional dysconnectivity observed in these disorders. Optical disector was used to estimate the numerical density (Nv) of oligodendrocytes and oligodendrocyte clusters (OLC) in the CN of cases with schizophrenia, BPD and major depressive disorder (MDD) and in normal controls (15 cases per group). A significant reduction in the Nv of oligodendrocytes was found in schizophrenia and BPD as compared to the control group (p < 0.05), and the Nv of OLC was significantly lowered in schizophrenia and BPD compared to controls (p < 0.05). There were no significant differences between MDD and control groups. The Nv of OLC was significantly decreased in the left hemisphere in schizophrenia as compared to the left hemisphere of the control group (-52%, p < 0.01). The data indicates that a decreased density of oligodendrocytes and OLC could contribute to the altered functional connectivity of the CN in subjects with severe mental illnesses.

Is there section deformation resulting in differential change of nuclear numerical densities along the z axis of thick methacrylate or paraffin sections?

The optical disector, a three-dimensional counting frame or probe in stereology, is often positioned in the middle (depth) of a thick section for unbiased nuclear counting. Using 30-40 µm thick methacrylate or paraffin sections for nuclear counting of neurons with the optical disector, however, some studies showed markedly higher nuclear densities at 10% of the section thickness near the top or bottom surface of the section, suggestive of deformation of section along its z axis and thus affecting the number estimation. To verify the findings, this study obtained two sets of 12-14 methacrylate sections (average thicknesses 21.7 and 29.4 µm) and two sets of 12 paraffin sections (average thicknesses 13.8 and 29.2 µm) from mature rat testes. Each section was used to count round spermatid nuclei in the seminiferous epithelium densely packed with the cells, using 3-4 consecutive disectors placed vertically (along the z axis of the section) from the top surface of the section, through the whole section thickness (two sets of methacrylate and paraffin sections) or in 80-83% of the thickness (other sections). The results demonstrated that, overall, there were no considerable nonuniform changes of the nuclear densities along the z axis of the sections.

Reduced oligodendrocyte density in layer 5 of the prefrontal cortex in schizophrenia.

Neuroimaging and post-mortem studies have implicated altered myelin integrity and oligodendrocyte abnormalities in the dysfunction of neuronal network in schizophrenia, including the prefrontal cortex, Brodmann area (BA) 10. Pyramidal neurons in layer 5 of BA10 are the important link of reciprocal frontal cortical-basal ganglia-thalamic circuits altered in schizophrenia. Previously, we found ultrastructural dystrophic and degenerative alterations of oligodendrocytes in layer 5 of BA10 in schizophrenia. The aim of the study was to estimate the numerical density (Nv) of oligodendrocytes in layer 5 of BA10 in schizophrenia as compared to normal controls. 17 chronic schizophrenia subjects and 22 healthy matched controls were studied in Nissl-stained sections using optical disector method. Group differences were analyzed using ANCOVA followed by post hoc Duncan's test. The Nv of oligodendrocytes was significantly lower (- 32%, p < 0.001) in the schizophrenia group as compared to the control group. Young controls (age < 50 years old) showed significantly higher Nv of oligodendrocytes as compared to elderly controls (age > 50 years old). Young and elderly schizophrenia subgroups did not differ significantly. Both control subgroups have significantly higher Nv of oligodendrocytes as compared to the schizophrenia subgroups. Decreased Nv of oligodendrocytes found in layer 5 of BA10 may be the result of dystrophic and destructive alterations and/or disrupted development of oligodendrocytes in schizophrenia.

Metformin attenuates increase of synaptic number in the rat spinal dorsal horn with painful diabetic neuropathy induced by type 2 diabetes: a stereological study.

In our previous study, we have shown that number of synapses in the L5 segment of spinal dorsal horn increased significantly in a rat model of painful diabetic neuropathy (PDN) induced by high-dose of streptozotocin (an animal model of type 1 diabetes). The aims of this study were: (1) to determine whether high fat diet/low dose streptozotocin-diabetes, a rat model for type 2 diabetes, related PDN was also associated with this synaptic plasticity, (2) to reveal the range of this synaptic plasticity change occurred (in the whole length of spinal dorsal horn or only in the L5 lumbar segment of spinal dorsal horn) and (3) to discover whether treatment with metformin had effect on this synaptic plasticity. Male adult Sprague-Dawley rats were randomly allocated into the control group (n = 7), the PDN group (n = 6) and the PDN treated with metformin (PDN + M) group (n = 7), respectively. 28 days after medication, synaptic and neuronal numbers in the whole length of spinal dorsal horn or in 1 mm length of the L5 segment of spinal dorsal horn were estimated by the optical disector (a stereological technique). Compared to the control group and the PDN + M group, number of synapses in the L5 segment of spinal dorsal horn increased significantly in the PDN group (P < 0.05). There was no significant change between the control group and the PDN + M group in terms of the parameters in the L5 segment of the spinal dorsal horn (P > 0.05). Parameters of the whole length of spinal dorsal horn showed no significant changes (P > 0.05). Our results suggest that high fat diet/low dose streptozotocin diabetes related PDN is also associated with a numerical increase of synapses in the L5 segment of spinal dorsal horn but not in the whole length of spinal dorsal horn. Furthermore, the analgesic effect of metformin against PDN is related to its inhibition of numerical increase of synaptic number in the rat spinal dorsal horn.

Absolute number of parvicellular and magnocellular neurons in the red nucleus of the rat midbrain: a stereological study.

The absolute number of parvicellular and magnocellular neurons in the red nucleus was estimated using design-based stereological counting methods and systematic random sampling techniques. Six young adult male rats, and a complete set of serial 40-µm glycolmethacrylate sections for each rat, were used to quantify neuronal numbers. After a random start, a systematic subset (i.e. every third) of the serial sections was used to estimate the total volume of the red nucleus using Cavalieri's method. The same set of sampled sections was used to estimate the number of neurons in a known subvolume (i.e. the numerical density Nv ) by the optical disector method. Multiplication of the total volume by Nv yielded the absolute number of neurons. It was found that the right red nucleus consisted, on average, of 8400 parvicellular neurons (with a coefficient of variation of 0.16) and 7000 magnocellular neurons (0.12). These total neuronal numbers provide important data for the transfer of information through these nuclei and for species comparisons.

Neurostereology protocol for unbiased quantification of neuronal injury and neurodegeneration.

Neuronal injury and neurodegeneration are the hallmark pathologies in a variety of neurological conditions such as epilepsy, stroke, traumatic brain injury, Parkinson's disease and Alzheimer's disease. Quantification of absolute neuron and interneuron counts in various brain regions is essential to understand the impact of neurological insults or neurodegenerative disease progression in animal models. However, conventional qualitative scoring-based protocols are superficial and less reliable for use in studies of neuroprotection evaluations. Here, we describe an optimized stereology protocol for quantification of neuronal injury and neurodegeneration by unbiased counting of neurons and interneurons. Every 20th section in each series of 20 sections was processed for NeuN(+) total neuron and parvalbumin(+) interneuron immunostaining. The sections that contain the hippocampus were then delineated into five reliably predefined subregions. Each region was separately analyzed with a microscope driven by the stereology software. Regional tissue volume was determined by using the Cavalieri estimator, as well as cell density and cell number were determined by using the optical disector and optical fractionator. This protocol yielded an estimate of 1.5 million total neurons and 0.05 million PV(+) interneurons within the rat hippocampus. The protocol has greater predictive power for absolute counts as it is based on 3D features rather than 2D images. The total neuron counts were consistent with literature values from sophisticated systems, which are more expensive than our stereology system. This unbiased stereology protocol allows for sensitive, medium-throughput counting of total neurons in any brain region, and thus provides a quantitative tool for studies of neuronal injury and neurodegeneration in a variety of acute brain injury and chronic neurological models.

Distribution of Particles in the Z-axis of Tissue Sections: Relevance for Counting Methods.

The distribution of particles in the z-axis of thick tissue sections has gained considerable attention, primarily because of implications for the accuracy of modern stereological counting methods. Three major types of artifacts can affect these sections: loss of particles from the surfaces of tissue sections (lost caps), homogeneous collapse in the z-axis, and differential deformation in the z-axis. Initially it was assumed that thick sections were not compromised by differential shrinkage or compression (differential uniform deformation). Studies in the last decade showed that such artifacts are common and that they depend on embedding media and sectioning devices. Paraffin, glycolmethacrylate and vibratome sections are affected by this artifact, but not celloidin sections or cryostat-derived cryosections. Differential distribution of particles in the z-axis is likely due to compression of the surface areas (margins) during sectioning, resulting in differential particle densities in the core and margin of tissue sections. This deformation of tissue sections can be rapidly assessed by measuring the position of particles in the z-axis. The analysis is complicated by potential secondary effects on section surfaces through loss of particles, the so-called "lost caps" phenomenon. Secondary effects necessitate the use of guard spaces, while their use in case of primary effects (compression due to sectioning) would enhance the artifact's impact on bias. Symmetric versus asymmetric patterns of z-axis distortion can give clues to distinguish primary and secondary effects. Studies that use the optical disector need to take these parameters into account to minimize biases.

Disector Z-axis mechanical method for stereology

The purpose of this work is to calibrate a not expensive microscope to be applied in Optical Disector estimation. The evaluation of "Z-axis" bias and the "Z-axis" calibration were made utilizing a cover slip and a "manual digimatic outside micrometer scale" (Mitutoyo, Japan). Calibrating the cover slip we performed the calibration of the "Z-axis" of a microscope. In the cover slip two lines were painted with different colors using a pen glass. A blue line was painted on the up surface and another line (red) on the bottom surface of the cover slip forming a cross. Two metal rings with one palette welded in each were adapted in the microscope. Other palette was welded in the gross focuses in order to restrict the route of the fine focuses (Z-axis study) using the two palettes fixed in both rings. Results show that 10 micrometers in "Z-axis" were equal to 3.2 micrometers in the scale of its fine focuses of the microscope. Then, a Disector of 10μm (10μm in Z-axis) is equal to 3.2 micrometers in the fine focuses of this microscope. In conclusion, "Z-axis" calibration is crucial to assure enough precision for Disector Method since the equipment can be manufactured without the ideal precision or its precision can be lost after use.

O propósito deste trabalho é a Calibração de um microscópio comum de baixo custo de forma a usá-lo para estimativas do Método Disector Óptico. A avaliação do viés e a calibração do "eixo-Z" foram feitas usando uma lamínula e um "micrômetro manual digital" (Mitutoyo, Japan). Com a cali-bração da lamínula, calibramos o "eixo-Z" de um microscópio. Foram pintadas duas linhas com diferentes cores utilizando canetas para vidro. Uma linha (azul) foi pintada na superfície superior da lamínula e outra linha (vermelha) foi pintada na superfície inferior da lamínula, formando uma cruz. Dois anéis de metal com uma paleta de metal soldada a cada um deles foram adaptados ao micrométrico do microscópio. Outra paleta foi soldada ao macrométrico do mesmo microscópio, o que foi feito de forma a limitar o percurso do micrométrico (eixo-Z) pela ação desta em conjunto com a ação das duas paletas fixadas no micrométrico. Os resultados mostraram que 10 micrômetros no "eixo-Z" do microscópio foi igual a 3,2 na escala do seu micrométrico. Assim, um Disector de 10μm (10μm no eixo-Z) é igual a 3,2 micrômetros no micrométrico do microscópio utilizado. Concluindo, a calibração do eixo-Z é crucial para assegurar a precisão suficiente para as avaliações usando o Método Disector, já que os equipamentos podem ser manufaturados sem a precisão ideal ou podem perder sua precisão durante o tempo com o uso.