Design-based Stereology Research Articles

Design‐based stereological estimation of hepatocyte number, by combining the smooth optical fractionator and immunocytochemistry with anti‐carcinoembryonic antigen polyclonal antibodies

Hepatocytes (HEP) have been the major target for structural quantification in the liver, but an estimation of their total number (N), their percentage in relation to the global number of liver cells and the evaluation of the percentage of binucleated hepatocytes (BnHEPs) have never been performed with modern design-based stereological techniques. The establishment of sound technical guidelines and baseline quantitative data in non-pathological conditions are relevant to properly evaluate HEP hyperplasia and BnHEP responses. In this study, we combined immunocytochemistry with sound design-based stereology for estimating the N of HEP and the N of non-hepatocytic cells (NHCs). For obtaining systematic uniform random sections (30 microm thick), a smooth fractionator sampling scheme was applied to the liver of five male Wistar rats (3 month old). Those sections were immunostained with polyclonal antibodies against carcinoembryonic antigen. Because biliary canaliculi were then marked, an unequivocal counting of mononucleated hepatocytes (MnHEP) and BnHEP was allowed. The N of HEP was estimated to be 1.93 x 10(9), with a coefficient of error (CE) of 0.02, corresponding to 129 x 10(6) HEP/g of liver. BnHEP represented 26% of total HEP number. The N of NHC was estimated as 1.31 x 10(9) (CE=0.02). The strategy here presented provides a reliable method for accessing the N of HEP (distinguishing MnHEP from BnHEP) in situations in which these parameters are relevant, namely for evaluating the magnitude of an hyperplastic liver response from its very early onset.

Efficient quantification of afferent cochlear ultrastructure using design-based stereology

The afferent synapse between the auditory nerve fiber and the inner hair cell (IHC) represents a critical junction for hearing. Elucidation of the structure at this site will help establish the substrate for normal sound encoding as well as pathologic processes associated with hearing dysfunction. Previous applications of unbiased (design-based) stereological principles have expanded our knowledge of neuro-morphological changes evident with the light microscope. Applying these principles at the level of the synapse is a promising morphometric approach for the efficient sampling of large reference spaces with electron microscopy. This study tests the accuracy of using ultra-thin sections at a fixed interval, known as disector pairs, to quantify afferent innervation density. We analyzed the total numbers of afferent terminals, synaptic thickenings, and synaptic bodies associated with each IHC in the C57BL/6J mouse cochlea, and confirmed the accuracy of the stereological approach in comparison to three-dimensional reconstructions of serial alternate sections. The higher sampling efficiency of the disector pair method rapidly increases precision while also reducing the largest source of variability, inter-animal differences. We conclude that ultrastructural quantification of afferent innervation can be accomplished in the cochlea using efficient design-based stereology.

Improved lung preservation relates to an increase in tubular myelin-associated surfactant protein A

BackgroundDeclining levels of surfactant protein A (SP-A) after lung transplantation are suggested to indicate progression of ischemia/reperfusion (IR) injury. We hypothesized that the previously described preservation-dependent improvement of alveolar surfactant integrity after IR was associated with alterations in intraalveolar SP-A levels.MethodsUsing immuno electron microscopy and design-based stereology, amount and distribution of SP-A, and of intracellular surfactant phospholipids (lamellar bodies) as well as infiltration by polymorphonuclear leukocytes (PMNs) and alveolar macrophages were evaluated in rat lungs after IR and preservation with EuroCollins or Celsior.ResultsAfter IR, labelling of tubular myelin for intraalveolar SP-A was significantly increased. In lungs preserved with EuroCollins, the total amount of intracellular surfactant phospholipid was reduced, and infiltration by PMNs and alveolar macrophages was significantly increased. With Celsior no changes in infiltration or intracellular surfactant phospholipid amount occurred. Here, an increase in the number of lamellar bodies per cell was associated with a shift towards smaller lamellar bodies. This accounts for preservation-dependent changes in the balance between surfactant phospholipid secretion and synthesis as well as in inflammatory cell infiltration.ConclusionWe suggest that enhanced release of surfactant phospholipids and SP-A represents an early protective response that compensates in part for the inactivation of intraalveolar surfactant in the early phase of IR injury. This beneficial effect can be supported by adequate lung preservation, as e.g. with Celsior, maintaining surfactant integrity and reducing inflammation, either directly (via antioxidants) or indirectly (via improved surfactant integrity).

Design–based stereological estimation of the total number of cardiac myocytes in histological sections

Counting the total number of cardiac myocytes has not previously been possible in ordinary histological sections using light microscopy (LM) due to difficulties in defining the myocyte borders properly. To describe a method by which the total number of cardiac myocytes is estimated in LM sections using design-based stereology. From formalin-fixed left rat ventricles (LV) isotropic uniformly random sections were cut. The total number of myocyte nuclei per LV was estimated using the optical disector. Two-microm-thick serial paraffin sections were stained with antibodies against cadherin and type IV collagen to visualise the intercalated discs and the myocyte membranes, respectively. Using the physical disector in "local vertical windows" of the serial sections, the average number of nuclei per myocyte was estimated. The total number of myocyte nuclei in LV was 34.1 x 10(6) (0.08) (mean (coefficient of variation)), and the mean number of nuclei per myocyte 1.85 (0.03). Combining these estimates the total number of myocytes in LV was calculated to be 18.5 x 10(6) (0.09). This new method is applicable to a range of experiments focusing on myocyte proliferation and death.

Selective Neuron Loss in the Paraventricular Nucleus of Hypothalamus in Patients Suffering from Major Depression and Bipolar Disorder

The structural basis for depressive disorders remains unknown. Studies using neuroimaging and postmortem brain tissue indicate that anatomic substrates may contribute to major depression disorder (MDD) and bipolar disorder (BD). The present study used design-based stereology to assess neuron loss in 2 well-defined hypothalamic structures, the paraventricular nucleus (PVN) and supraoptic nucleus (SON), in clinically well-studied cases with severe depression. The left or right diencephalon was blocked from 26 brains removed at autopsy from age-matched controls (5 male/3 female) and patients with MDD (6 male/5 female) and BD (5 male/2 female). Serial sections were cut at an instrument setting of 60 microm through the entire PVN and SON from left hypothalamus and 8 to 10 sections per brain were systematic-random sampled and stained with cresyl violet. A trained operator blind to clinical diagnosis used computerized stereology to estimate total neuron numbers in PVN and SON. The results revealed a selective, robust reduction of approximately 50% in total neuron number in the PVN for major depression and bipolar cases compared with age-matched controls, with no differences in neuron numbers in the SON. This selective neuronal loss in the PVN appears to identify an important neurobiologic substrate for the behavioral manifestations of depression.

Effect of prenatal protein malnutrition on numbers of neurons in the principal cell layers of the adult rat hippocampal formation

Malnutrition has been associated with a variety of functional and anatomical impairments of the hippocampal formation. One of the more striking of these is widespread loss of hippocampal neurons in postnatally malnourished rats. In the present study we have investigated the effect of prenatal malnutrition on these same neuronal populations, neurons that are all generated during the period of the dietary restriction. In prenatally protein deprived rats, using design-based stereology, we have measured the regional volume and number of neurons in the hilus of the dentate gyrus and the pyramidal cell layers of CA3, CA2, CA1, and the subiculum of 90-day-old animals. These results demonstrated a statistically significant reduction of 20% in neuron numbers in the CA1 subfield, while numbers in the other subfields were unchanged. There was a corresponding significant reduction of 22% in the volume of the CA1 subfield and a significant 14% decrease in the volume of the pyramidal layer of the subiculum. The change in volume of the pyramidal layer of the subiculum without neuron loss may reflect loss of CA1 afferent input to the pyramidal layer. Although the effect of nutritional deprivation on the neuronal population appears to be different in pre- and postnatal malnutrition, both dietary paradigms highlight the vulnerability of key components of the hippocampal trisynaptic circuit (consisting of the dentate granule cell mossy fibers projection to CA3 pyramids and the CA3 projection to the CA1 pyramids), which is an essential circuit for memory and learning.

Regional estimates of hair cells and supporting cells in the human crista ampullaris

Regional estimates of type I and type II vestibular hair cells (HC) and supporting cell (SC) numbers were obtained from the horizontal crista ampullaris by using design-based stereology in human. Cristae were microdissected from temporal bones obtained post-mortem (N=16, age range 26-98 years). Three groups were made according to age: group 1, n=5, ages between 26 and 67 years, average age 51 years; group 2, n=4, average age 84 years; and group 3, n=7, average age 94 years. For group 1, the average total HC number was 8,005+/-214, corresponding to 4,119+/-107 type I HC, 3,886+/-117 type II HC, and 10,274+/-224 SC. The type I:type II HC ratio was 1.06+/-0.01, and HC density was 0.80 cells/100 microm2. For group 2, the average total HC number was 7,074+/-489, corresponding to 3,733+/-212 type I HC, 3,341+/-314 type II HC, and 9,321+/-858 SC. The type I:II HC ratio was 1.12+/-0.06, and HC density was 0.75 cells/100 microm2. For group 3, the average HC number was 6,009+/-327, corresponding to 3,380+/-223 type I HC, 2,628+/-235 type II HC, and 10,185+/-182 SC. The type I:II HC ratio was 1.34+/-0.10, and HC density was 0.63 cells/100 microm2. A significant decline in type I, type II, and total HC number and density was found in groups 2 and 3, with individuals exceeding the average human life span.

Regional hippocampal alteration associated with cognitive deficit following experimental brain injury: A systems, network and cellular evaluation

Cognitive deficits persist in patients who survive traumatic brain injury (TBI). Lateral fluid percussion brain injury in the mouse, a model of human TBI, results in hippocampal-dependent cognitive impairment, similar to retrograde amnesia often associated with TBI. To identify potential substrates of the cognitive impairment, we evaluated regional neuronal loss, regional hippocampal excitability and inhibitory synaptic transmission. Design-based stereology demonstrated an approximate 40% loss of neurons through all subregions of the hippocampus following injury compared with sham. Input/output curves recorded in slices of injured brain demonstrated increased net synaptic efficacy in the dentate gyrus in concert with decreased net synaptic efficacy and excitatory postsynaptic potential-spike relationship in area CA1 compared with sham slices. Pharmacological agents modulating inhibitory transmission partially restored regional injury-induced alterations in net synaptic efficacy. Both evoked and spontaneous miniature inhibitory postsynaptic currents (mIPSCs) recorded in surviving dentate granule neurons were smaller and less frequent in injured brains than in uninjured brains. Conversely, both evoked and spontaneous mIPSCs recorded in surviving area CA1 pyramidal neurons were larger in injured brains than in uninjured brains. Together, these alterations suggest that regional hippocampal function is altered in the injured brain. This study demonstrates for the first time that brain injury selectively disrupts hippocampal function by causing uniform neuronal loss, inhibitory synaptic dysfunction, and regional, but opposing, shifts in circuit excitability. These changes may contribute to the cognitive impairments that result from brain injury.

Prenatal protracted irradiation at very low dose rate induces severe neuronal loss in rat hippocampus and cerebellum

Prenatal irradiation is known to damage the developing brain. However, little is known about the consequences of very low dose rate prenatal protracted irradiation over several days on neuron numbers in the offspring brain, and on volumes of the corresponding brain regions. Pregnant Wistar rats were exposed either to a protracted γ irradiation from embryonic day (E) 13 to E16 (0.7 mGy/min; total cumulative dose approximately 3 Gy) or were sham-irradiated. Thirty months old male and female offspring were then analyzed for alterations in hippocampal and cerebellar morphology. Using design-based stereology and the analysis of sets of sections systematically and randomly sampled to span the entire brain region of interest, a statistically significant decrease in numbers of hippocampal pyramidal and granule cells as well as of cerebellar Purkinje and granule cells (approximately 50%) was found in male and female irradiated offspring. The volumes of these brain regions were comparably altered. The analysis of only a “representative” section per animal yielded mostly non-significant trends. Evaluation of neuron densities showed no differences between prenatally irradiated and sham-irradiated offspring. Most importantly, very low dose rate prenatal protracted γ irradiation did not result in the same morphologic alterations in the offspring brain as previously observed after prenatal single irradiation such as derangement of the laminar structure of pyramidal cells within the hippocampus or malformation of cerebellar lobules.

Neuropathological findings in autism.

Autism is currently viewed as a largely genetically determined neurodevelopmental disorder, although its underlying biological causes remain to be established. In this review, we examine the available neuropathological literature on autism and discuss the findings that have emerged. Classic neuropathological observations are rather consistent with respect to the limbic system (nine of 14 studied cases showed increased cell packing density and smaller neuronal size), the cerebellum (21 of 29 studied cases showed a decreased number of Purkinje cells, and in all of five cases that were examined for age-related morphological alterations, these changes were found in cerebellar nuclei and inferior olive) and the cerebral cortex (>50% of the studied cases showed features of cortical dysgenesis). However, all reported studies had to contend with the problem of small sample sizes, the use of quantification techniques not free of bias and assumptions, and high percentages of autistic subjects with comorbid mental retardation (at least 70%) or epilepsy (at least 40%). Furthermore, data from the limbic system and on age-related changes lack replication by independent groups. It is anticipated that future neuropathological studies hold great promise, especially as new techniques such as design-based stereology and gene expression are increasingly implemented and combined, larger samples are analysed, and younger subjects free of comorbidities are investigated.

GM-CSF mediates alveolar epithelial type II cell changes, but not emphysema-like pathology, in SP-D-deficient mice.

Surfactant protein D (SP-D) is a member of the collectin subfamily of C-type lectins, pattern recognition proteins participating in the innate immune response. Gene-targeted mice deficient in SP-D develop abnormalities in surfactant homeostasis, hyperplasia of alveolar epithelial type II cells, and emphysema-like pathology. Granulocyte/macrophage colony-stimulating factor (GM-CSF) is required for terminal differentiation and subsequent activation of alveolar macrophages, including the expression of matrix metalloproteinases and reactive oxygen species, factors thought to contribute to lung remodeling. Type II cells also express the GM-CSF receptor. Thus we hypothesized GM-CSF might mediate some or all of the cellular and structural abnormalities in the lungs of SP-D-deficient mice. To test this, SP-D (D-G+) and GM-CSF (D+G-) single knockout mice as well as double knockout mice deficient for both SP-D and GM-CSF (D-G-) were analyzed by design-based stereology. Compared with wild type, D-G+ as well as D+G- mice showed decreased alveolar numbers, increased alveolar sizes, and decreased alveolar epithelial surface areas. These emphysema-like changes were present to a greater extent in D-G- mice. D-G+ mice developed type II cell hyperplasia and hypertrophy with increased intracellular surfactant pools, whereas D+G- mice had smaller type II cells with decreased intracellular surfactant pools. In contrast to the emphysematous changes, the type II cell alterations were mostly corrected in D-G- mice. These results indicate that GM-CSF-dependent macrophage activity is not necessary for emphysema development in SP-D-deficient mice, but that type II cell metabolism and proliferation are, either directly or indirectly, regulated by GM-CSF in this model.

The Use of Multiple Fluorescence Labels for Determination of Cell Number and Phenotype by Design-Based Stereology

Extended abstract of a paper presented at Microscopy and Microanalysis 2004 in Savannah, Georgia, USA, August 1–5, 2004.

Implementing Design-Based Stereology

Extended abstract of a paper presented at Microscopy and Microanalysis 2004 in Savannah, Georgia, USA, August 1–5, 2004.

A standardized method for brain-cutting suitable for both stereology and MRI-brain co-registration

We have developed an agar-embedding method for brain-slicing that minimizes the geometrical distortions which arise from handling and slicing the fixed postmortem brain. To facilitate postmortem brain-magnetic resonance imaging (MRI) co-registration, each hemisphere is processed separately. We embed the fixed brain hemisphere with reference markers in agar. The block containing the brain and markers is sliced at a fixed interval using a rotary slicer. Each slice is photographed with a high-resolution digital camera. The digital images are realigned as a 3-dimensional volume via a control point-based registration method for multi-slice registration. The realigned multiple slices of the reconstructed postmortem hemisphere are then co-registered to corresponding slices of an in vivo reference MRI-volume. We illustrate these postmortem MRI-brain co-registration methods to correlate in vivo T 2-weighted MRI hyperintensities in gray and white matter with underlying pathology. For design-based stereology, the volume of interest (VOI) is defined using reproducible anatomical boundaries. This method is suitable for stereologic measures of structures ranging from defined nuclei to whole brain.

The anatomy of the porcine subthalamic nucleus evaluated with immunohistochemistry and design-based stereology

This study provides a light-microscopic description of the organization, morphology and number of neurons in the subthalamic nucleus (STN) of the Göttingen minipig. It is based on histological material stained with Nissl, Golgi and autometallographic techniques, and employs design-based stereological estimation of the total neuron number. The organization of several neurotransmitters in the STN has been evaluated in histological preparations stained for acetylcholinesterase (AChE) and immunostained for choline acetyltransferase (ChAT), tyrosine hydroxylase (TH), glutamic acid decarboxylase (GAD) and glutamate. In all of the stained preparations the STN appeared as a distinct lens-shaped structure located in the caudal diencephalon, medial to the internal capsule and ventrolateral to the zona incerta. Rostrally, the STN approached the globus pallidus pars interna, whereas caudally the ventromedial part of the STN was adjacent to the rostral part of the substantia nigra pars compacta (SNc), where some of the neurons of the two nuclei merged. The neurons in the STN had medium-sized (25-40 microm) ovoid or fusiform cell bodies, from which three to six large dendrites emanated in a direction predominantly parallel to the long axis of the STN. Immunohistochemistry revealed that most of the subthalamic neurons were glutamatergic and differed significantly in appearance from the large stellate TH-positive cells of the adjacent SNc. Numerous TH-positive bouton-rich fibers traversed the STN. The GAD-staining revealed a large number of terminals within the boundaries of the STN. The STN was highly AChE-positive, reflecting a prominent innervation by ChAT-positive terminals. The total number of subthalamic neurons in one hemisphere was estimated to be approximately 56,000. We conclude that the neuroarchitecture of the porcine STN is similar to primates, including humans, and appears well-suited for further studies examining the role of the STN in movement disorders.

Effects of estrogen and raloxifene on neuroglia number and morphology in the hippocampus of aged female mice

Hormone replacement therapy with the gonadal steroid estrogen or synthetic agents such as raloxifene, a selective estrogen receptor modulator, may affect cellular function in brains of postmenopausal women. In vitro studies suggest that 17β estradiol and raloxifene can alter the microglial and astrocyte expression of immuno-neuronal modulators, such as cytokines, complement factors, chemokines, and other molecules involved in neuroinflammation and neurodegeneration. To directly test whether exogenous 17β estradiol and raloxifene affect the number of glial cells in brain, C57BL/6NIA female mice aged 20–24 months received bilateral ovariectomy followed by s.c. placement of a 60-day release pellet containing 17β estradiol (1.7 mg), raloxifene (10 mg), or placebo (cholesterol). After 60 days, numbers of microglia and astrocytes were quantified in dentate gyrus and CA1 regions of the hippocampal formation using immunocytochemistry and design-based stereology. The results show that long-term 17β estradiol treatment in aged female mice significantly lowered the numbers of astrocytes and microglial cells in dentate gyrus and CA1 regions compared with placebo. After long-term treatment with raloxifene, a similar reduction was observed in numbers of astrocytes and microglial cells in the hippocampal formation. These findings indicate that estrogen and selective estrogen receptor modulators can influence glial-mediated inflammatory pathways and possibly protect against age- and disease-related neuropathology.

Surfactant homeostasis is maintained in vivo during keratinocyte growth factor-induced rat lung type II cell hyperplasia.

Keratinocyte growth factor (KGF) induces transient proliferation of alveolar type II cells (AEII) associated with surfactant alterations. To test the hypothesis that homeostasis of intracellular phospholipid stores is maintained under KGF-induced hyperplasia, we (1) collected tissue from adult rat lungs, fixed for light and electron microscopy 3 days after intratracheal instillation of 5 mg recombinant human (rHu) KGF/kg body weight or phosphate-buffered saline (PBS), and from untreated control animals (five animals/group) for design-based stereology of AEII and lamellar body (LB) ultrastructure; and (2) we analyzed uptake and distribution of instilled radiolabeled phospholipids. After rHuKGF, AEII-coverage of alveolar walls (PBS:8.3 +/- 3.0%; rHuKGF:30.6 +/- 4.8%) and number of AEII/ml lung volume (PBS:28.5 +/- 6.5 x 10(6); rHuKGF:48.2 +/- 5.8 x 10(6)) were increased (p < 0.008). Number (PBS:97 +/- 25; rHuKGF:54 +/- 7) and volume (PBS:45.3 +/- 13.8 microm(3); rHuKGF:21.0 +/- 4.7 microm(3)) of LBs per cell were decreased (p < 0.008), but not total amount/ml lung volume (PBS:128 +/- 46. 4 x 10(7) microm(3); rHuKGF:103 +/- 34. 7 x 10(7) microm(3)). This was paralleled by a shift to larger LBs. After rHuKGF, radiolabeled phospholipids accumulated in whole lung tissue relative to lavage fluid (p < 0.01). However, less radiolabel was incorporated per cell (p < 0.01). We conclude that under rHuKGF-induced AEII proliferation intracellular surfactant was decreased per single cell, whereas a constant amount was maintained per unit lung volume. We suggest that surfactant homeostasis is regulated at the level of phospholipid transport processes, for example, secretion and reuptake.

Number of Langerhans immune cells in painful and non-painful human skin after shingles.

During injury or inflammation, paracrine sensitization of peripheral sensory neurons by immune cells contributes to the sensation of pain. It is less clear whether this neural sensitization contributes to neuropathic pain after neural injury as well. Shingles (herpes zoster) is a common disease that leaves some patients with prolonged neuropathic pain known as postherpetic neuralgia (PHN). Sensitization of cutaneous neurons has been hypothesized to contribute to PHN. Langerhans cells (LC), the Ia(+) macrophages of the skin, contact epidermal neurites and, when activated, synthesize molecules with the ability to sensitize axons. For these reasons, we examined morphological evidence for activation of LC in subjects with established PHN. We also evaluated the relationship between numbers of LC and nociceptive epidermal nerve endings; these are markedly reduced in PHN. We used design-based stereology to estimate the number of CD1a(+) LC in biopsies of painful and non-painful skin from ten adults with or without PHN after shingles on the torso. There were no differences in the number of LC in previously shingles-affected and normal-control skin biopsies. The number of LC also remained at normal levels in biopsies with near-loss of innervation from shingles. LC numbers were unrelated to the presence or severity of pain. These data suggest that neuropathic pain in established PHN is not associated with increased numbers of cutaneous macrophages, and that the number of cutaneous macrophages in skin from the human torso is independent of the number of epidermal nerve endings.

A note on design-based versus model-based variance estimation in stereology

Recently, systematic sampling on the circle and the sphere has been studied by Gual-Arnau and Cruz-Orive (2000) from a design-based point of view. In this note, it is shown that their mathematical model for the covariogram is, in a model-based statistical setting, a special case of the p-order shape model suggested by Hobolth, Pedersen and Jensen (2000) and Hobolth, Kent and Dryden (2002) for planar objects without landmarks. Benefits of this observation include an alternative variance estimator, applicable in the original problem of systematic sampling. In a wider perspective, the paper contributes to the discussion concerning design-based versus model-based stereology.

A note on design-based versus model-based variance estimation in stereology

Recently, systematic sampling on the circle and the sphere has been studied by Gual-Arnau and Cruz-Orive (2000) from a design-based point of view. In this note, it is shown that their mathematical model for the covariogram is, in a model-based statistical setting, a special case of the p-order shape model suggested by Hobolth, Pedersen and Jensen (2000) and Hobolth, Kent and Dryden (2002) for planar objects without landmarks. Benefits of this observation include an alternative variance estimator, applicable in the original problem of systematic sampling. In a wider perspective, the paper contributes to the discussion concerning design-based versus model-based stereology.