ReviewDesign-based stereology in neuroscience
Section snippets
Contents
Section 1: Basic information about design-based stereology in the literature
Section 2: Considerations for specimen preparation for design-based stereologic analysis
Section 3: Laboratory equipment for design-based stereologic analyses
Section 4: Potential bias in results of design-based stereologic analyses
Section 5: Parameters that can be assessed by design-based stereology
5.1: Considerations about delineation of brain regions.
5.2: Volume of a brain region
5.3: Number of neurons within a given
Section 1: basic information about design-based stereology in the literature
More than 20 reviews and eight comprehensive books dedicated to design-based stereology and its applications in neurosciences are available. Among the reviews, some address more basic concepts of design-based stereology, whereas others deal with specific details of this methodology. Within the first category, the papers by Gundersen (1986, 1992), Cruz-Orive and Weibel (1990), Coggeshall and Lekan (1996), Mayhew and Gundersen (1996) and Royet (1991) provide introductions to the basic concepts of
Section 2: considerations for specimen preparation for design-based stereologic analysis
Design-based stereologic methods have been developed to make statements about structures such as an identified organ, a definable brain region, a population of cells, or linear biological structures within a tissue. In the case of the brain, if such statements are to be valid for an entire brain region, then the analyzed sample of sections and microscopic fields has to be representative of it. This requires that one has access to the entire brain region, that all cells or linear biological
Section 3: laboratory equipment for design-based stereologic analyses
Recent developments in design-based stereology include techniques and analyses that have become possible only with the introduction of computer-interfaced microscopes and imaging instrumentation, in particular, estimates of local volumes with the Optical Rotator (section 5.5), estimates of the length of linear biological structures with Space Balls (section 5.7), and investigations on the spatial distribution of cells with Nearest-Neighbor analysis (section 5.8). Furthermore, the availability
Section 4: potential bias in results of design-based stereologic analyses
Let us assume a stereologic estimate of the total number of neurons within a certain brain region. If this estimate could be repeated ad infinitum and the mean of the estimates would equal the (unknown) true total number of neurons within this region, the estimates would be unbiased (i.e. without systematic error; for details see West, 2002). Importantly, the use of design-based stereology does not guarantee unbiasedness of the corresponding estimates (Guillery and Herrup, 1997). Rather, the
5.1: considerations about delineation of brain regions
Each stereologic investigation starts with the identification of the boundaries of the brain region of interest (thereafter abbreviated as BROI) on a systematic-random series of sections throughout this BROI (Fig. 1A). In many cases (such as the cerebellar granule cell layer) the boundaries of the BROI can easily be identified on Nissl-stained sections, and one can trace the boundaries on video images displayed on a computer (Glaser and Glaser, 2000). Alternatively, one can perform only a rough
Section 6: variability of estimates obtained with design-based stereology
The results of quantitative histologic investigations performed with design-based stereologic methods are estimates rather than exact measurements. Thus, the results will vary if the same stereologic estimate were to be independently repeated. This implies that information about the variability of stereologic estimates is an important topic.
If the same stereologic estimate would be repeated ad infinitum, one could calculate the coefficient of variation of these estimates (note that the mean of
Section 7: presentation of results obtained with design-based stereology
Investigators and readers, as well as editorial boards and reviewers have all been concerned by the issue of what information and discussion are essential in a stereologic paper. Whereas there is no general answer to this question, the following recommendations might be useful.
The materials and methods of stereologic studies should provide a description of the histologic processing, with consideration of the specific requirements to be met (outlined in section 2) and the potential sources of
Section 8: the future of design-based stereology in neuroscience
Over the recent years design-based stereology has become the state-of-the-art methodology in quantitative histologic analyses, and the application of design-based stereologic methods to the analysis of the CNS has considerably contributed to our understanding of the functional and pathological morphology of the brain. Nevertheless, design-based stereology is still an evolving field and it can be expected that its applications in neuroscience will be subject to many improvements and extensions
Concluding remarks
Many important questions in neuroscience cannot simply be answered with a simple qualitative analysis. Whenever quantitative morphologic approaches are used, the following issues should be considered: are the results representative for the entire BROI? Are the results independent from the size, shape, spatial orientation, and spatial distribution of the cells to be investigated? Is bias (i.e. systematic error) in the results kept to a minimum? Is it possible to assess the variability of the
Acknowledgments
We thank our many colleagues who prompted us to write this review. We would like to thank the Stanley Medical Research Institute and the US-National Alliance of Autism Research (to C.S. and P.R.H.), the Alzheimer Forschung Initiative e.V., the Internationale Stichting Alzheimer Onderzoek and the Hersenstichting (to C.S.), and NIH grants AG02219, AG05138, MH58911, and MH66392 (to P.R.H.). P.R.H. is the Regenstreif Professor of Neuroscience.
References (120)
Assaying structural changes after nerve damage, an essay on quantitative morphology
Pain Suppl
(1999)Research uses of neuropathological data
Neurobiol Aging
(1997)- et al.
Voronoi tessellation to study the numerical density and the spatial distribution of neurones
J Chem Neuroanat
(2000) - et al.
Evaluation of neuronal numerical density by Dirichlet tessellation
J Neurosci Methods
(1994) - et al.
Differential tissue shrinkage and compression in the z-axisimplications for optical disector counting in vibratome-, plastic- and cryosections
J Neurosci Methods
(2003) - et al.
Comparison of MR imaging against physical sectioning to estimate the volume of human cerebral compartments
Neuroimage
(2003) - et al.
Stereology, morphometry, and mappingthe whole is greater than the sum of its parts
J Chem Neuroanat
(2000) History of neuromorphometry
J Neurosci Methods
(1986)- et al.
Celloidin mounting (embedding without infiltration)a new, simple and reliable method for producing serial sections of high thickness through complete human brains and its application to stereological and immunohistochemical investigations
J Chem Neuroanat
(2000) - et al.
Loss and altered spatial distribution of oligodendrocytes in the superior frontal gyrus in schizophrenia
Biol Psychiatry
(2003)