Electronic Supplementary Material
Specimens, preparation, and immunohistochemical staining
The majority of specimens came from adults, with the exception of three juveniles that had a brain size comparable to their species typical adult average (Trachypithecus francoisi, Ateles geoffroyi, and Pithecia pithecia). Although some of the brains came from older individuals, it is important to note that the human, chimpanzee, and macaque monkey samples, which were used for the majority of phenotypic analyses, were all from non-geriatric adults. None of the brains showed evidence of neuropathological abnormality on routine examination. Most nonhuman primate postmortem brain specimens used in this study were provided by zoological or research institutions subsequent to immersion fixation in 10% buffered formalin after a variable postmortem interval (PMI), which never exceeded 14 hours. In addition, some brains were obtained from animals that were perfused transcardially with 4% paraformaldehyde in the context of unrelated experiments (Papio anubis, Macaca maura, Erythrocebus patas, Saimiri boliviensis, Aotus trivirgatus, Ateles geoffroyi). Brains came from animals that were housed in accordance with each institution’s guidelines. The golden guenon monkey brain was provided by the Office Rwandais du Tourisme et des Parcs Nationaux and the Mountain Gorilla Veterinary Project in compliance with CITES regulations. Human brain specimens were provided by Northwestern University Alzheimer’s Disease Center Brain Bank (3 women, 3 men, age range 35-54 years). For human cases, the PMI prior to formalin fixation ranged from 6 to 17 hours. The human cases showed no evidence of dementia before death and all individuals received a score of 0 for the CERAD senile plaque grade (Mirra et al. 1991) and the Braak and Braak (1991) neurofibrillary tangle stage. In all cases, brains were transferred to 0.1 M phosphate buffered saline (PBS, pH 7.4) containing 0.1% sodium azide and stored at 4˚C as soon as possible after acquisition to mitigate excessive antigen blockage and tissue shrinkage.
Brain mass of nonhuman specimens was measured either immediately after perfusion or directly upon receipt from the donating institution. Human brain masses were measured at autopsy and were provided by the brain bank. Some artifact in our total brain mass data due to interindividual differences in fixation length was unavoidable. Nevertheless, shrinkage artifact is likely to be minimal relative to the variance among species because our brain mass measurements fall within the normal range of values reported in the literature from fresh brains and in vivo MRI volumes measured from these taxa (Rilling & Insel 1999; Stephan et al. 1981; Zilles & Rehkämper 1988).
For all nonhuman specimens, the entire left frontal lobe was sectioned in the coronal plane. Human brain samples were dissected by the donating brain bank from the regions of interest in the left hemisphere as 4 cm-thick slabs. Tissue blocks were cryoprotected by immersion with increasing concentrations of sucrose solution up to 30%. Blocks were then frozen on a slur of dry ice and isopentane, and sections were cut at 40 μm using a sliding microtome.
Prior to immunostaining, sections were rinsed thoroughly in PBS, pretreated for antigen retrieval by incubation in 10 mM sodium citrate buffer (pH 3.5) at 37° C in an oven for 30 minutes, then immersed in a solution of 0.75% hydrogen peroxide in 75% methanol to eliminate endogenous peroxidase activity. Primary antibodies were diluted in a solution containing PBS with 2% normal serum and 0.1% Triton X-100 and incubated for approximately 48 hours on a rotating table at 4° C. After rinsing in PBS, sections were incubated in the secondary antibody (either biotinylated anti-mouse IgG or biotinylated anti-rabbit IgG, dilution 1:200: Vector Laboratories, Burlingame, CA) and processed with the avidin-biotin-peroxidase method using a Vectastain Elite ABC kit (Vector Laboratories). Immunoreactivity was visualized using 3,3’-diaminobenzidine (DAB) as a chromogen and intensified with nickel in some cases. For every specimen, specificity of the immunoreaction was confirmed by processing control sections as described above excluding the primary antibody. Immunostaining was completely absent in control sections. Only sections with clear staining of neuronal perikarya and dendrites were included in quantitative analyses. For each immunostained series, alternate sections were counterstained with cresyl violet to visualize non-immunoreactive neurons, cytoarchitectural boundaries, and cortical layers.
Stereologic analyses
Quantification of cellular densities within layers II-III was performed using a computerized stereology system consisting of a Zeiss Axioplan 2 photomicroscope equipped with a Ludl XY motorized stage, Heidenhain z-axis encoder, an Optronics MicroFire color videocamera, a Dell PC workstation, and StereoInvestigator software (MBF Bioscience, Wiliston, VT). Beginning at a random starting point, three equidistantly spaced sections were selected for analysis for each neuron type. Only the central portion of each cortical region was used for analyses, avoiding the boundaries at the transition between cytoarchitectonic areas. Optical disector frames (30 µm × 30 µm for Nissl stain; 65 µm × 65 µm for calcium-binding proteins) were placed in a systematic random fashion to cover the region of interest with approximately 30 frames per section, and the exact spacing of disectors depending on the size of the brain. Disector analysis was performed under Koehler illumination using a 63× objective (Zeiss Plan-Apochromat, N.A. 1.4). In immunostained sections, only interneurons that showed dark staining, a clear soma, and proximal dendritic segments were counted. Cells were counted when their centroid was encountered within the optical disector frame according to standard stereologic principles (Howard & Reed 1998; Mouton 2002). The thickness of optical disectors was set to 6 µm to allow for a minimum 2-µm guard zone on either side of the section after z-axis shrinkage from histological processing. Neuron numerical densities (Nv) were derived from these optical disector counts and corrected by the number-weighted mean section thickness as described in Sherwood et al. (2007).
Area 9 of dorsolateral prefrontal cortex (DLPFC) was selected for our broad interspecific analyses because it has been described as anatomically homologous in a diverse range of anthropoids, e.g., Callithrix jacchus (Burman et al. 2006), Saimiri oerstedii (Rosabal 1967), Macaca sp. (Petrides & Pandya 1999; von Bonin & Bailey 1947; Walker 1940), Papio hamadryas (Watanabe-Sawaguchi et al. 1991), Pan troglodytes (Bailey et al. 1950), Homo sapiens (Petrides & Pandya 1999; Rajkowska & Goldman-Rakic 1995). Cortical areas 4, 9, 32, and 44 in humans, chimpanzees, and macaque monkeys were identified based on their characteristic cytoarchitecture using descriptions from previous parcellations of these areas (Bailey et al. 1950; Paxinos et al. 2000; Petrides et al. 2005; Petrides & Pandya 1999; Sherwood et al. 2003; Sherwood et al. 2004; Sherwood et al. 2006; Vogt et al. 1995).
Consideration of fixation conditions and postmortem interval
Given the opportunistic nature of the brain collection for this study, it is important to consider the effect of fixation condition and postmortem interval (PMI) on the quantitative results. We examined directly whether fixation condition had a significant effect on the immunostaining quality of neurons in our sample by performing a restricted comparison among Old World monkey brains that had been prepared by immersion-fixation (n = 6) and perfusion-fixation (n = 10). We selected the Old World monkeys for this comparison because this phylogenetic group contained the most equal representation of brains that were fixed by each method. The percentage of the total layer II-III neuron population was calculated for each interneuron subtype in DLPFC and evaluated for differences between fixation conditions using independent samples t tests. Results showed no significant differences in the percentage of interneuron subtypes based on fixation method (CB: t = -0.25, P = 0.81; CR: t = 1.14, P = 0.30; PV: t = 1.61, P = 0.15). Thus, potential artifact in immunostaining against calcium-binding proteins from immersion-fixation with a 14-hour PMI is statistically indistinguishable from perfused tissue.
To further assess whether the length of PMI affected the quality of immunohistochemical staining, nonparametric Spearman’s correlation coefficients were calculated for PMI and the percentage of each interneuron subtype in each cortical area in humans. Data on specific PMI were not available for the other species in the sample. Out of twelve possible comparisons, there were two significant correlations with PMI (CB-ir interneurons in area 9: rs = 0.81, P = 0.05; CR-ir interneurons in area 44: rs = -0.81, P = 0.05); however these correlations were in opposite directions and neither was significant after applying a sequential Bonferroni correction of α for multiple tests. It is notable that Lavenex and colleagues (2009) also did not find a difference in the number of neurons immunoreactive for CB, CR, and PV in the hippocampal formation of rhesus monkeys when comparing brains that were paraformadehyde-fixed by perfusion versus immersion with varying PMIs extending as long as 48 hours.
We also used Spearman’s rank order correlations to test whether there was a relationship between age at death and the percentage of each interneuron subtype in each cortical area within macaques, chimpanzees, or humans. Out of 36 possible comparisons, there was only one significant correlation with age (PV-ir interneurons in area 44 of macaques: rs = -0.85, P = 0.03); however this correlation was not significant after Bonferroni correction of α. Thus, we conclude that PMI and age did not contribute significantly to the observed immunostaining patterns of interneurons.
Statistical analyses
For reduced major axis (RMA) scaling analyses and regression predictions, we used total neuron density as the independent variable to compare with interneuron subtype densities. In such bivariate plots, part of the independent variable is comprised by the dependent variable because interneurons contribute to the total neuron density. Some authors have raised the concern that this part-whole relationship may be particularly problematic for dependent variables that constitute a large proportion of the independent variable. When this is not taken into account, autocorrelation artificially inflates the correlation between variables and reduces the sensitivity of the regression to detecting departures from allometric expectations (Deacon 1990; Holloway 1979). To examine the possibility that such effects were present in our data, we re-analyzed each scaling relationship for interneuron densities versus total neuron density by subtracting the dependent value from the independent value. All relationships retained essentially the same patterns of statistical significance, coefficients of determination, and scaling exponents as when the analyses were performed with total neuron density as the independent variable.
For independent contrasts analyses, alternative methods of branch length transformation did not significantly alter the results and independent contrasts were uncorrelated with their standard deviations, indicating that branch lengths meet statistical assumptions (Garland et al. 1992).
Quantifying the partitioned variation
Treating each interneuron subtype separately, the first step in variance partitioning requires the calculation of fraction a+b, which is obtained from the regression of the predictor variable on the dependent variable. The resultant coefficient of determination (r2) is entered as the fraction a+b and represents the total uncorrected variance attributed to the undifferentiated components representing the interaction of phylogeny and the predictor. Step 2 is the calculation of fraction b+c by computing a multiple regression of the dependent variable on principal coordinates, which are derived from a distance matrix representing the phylogeny. Step 3 is the determination of fraction a+b+c by computing a multiple regression of the dependent variable on the predictor variable and the principal coordinates representing the phylogeny. Step 4 is the calculation of the decomposed component values for a, b, c, and d (i.e., unexplained variance) by subtraction from previous results, e.g., a = r2 (Step 3) – r2 (Step 2); c = r2 (Step 3) – r2 (Step 1). All multiple regressions were calculated using SPSS version 11.0 whereas the resultant distance matrix was calculated using Compare 2.0.
References
Bailey, P., von Bonin, G. & McCulloch, W. S. 1950 The Isocortex of the Chimpanzee. Urbana, IL: University of Illinois Press.
Braak, H. & Braak, E. 1991 Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82, 239-59.
Burman, K. J., Palmer, S. M., Gamberini, M. & Rosa, M. G. 2006 Cytoarchitectonic subdivisions of the dorsolateral frontal cortex of the marmoset monkey (Callithrix jacchus), and their projections to dorsal visual areas. J Comp Neurol 495, 149-72.
Deacon, T. W. 1990 Fallacies of progression in theories of brain-size evolution. International Journal of Primatology 11, 193-235.
Garland, T., Harvey, P. H. & Ives, A. R. 1992 Procedures for the analysis of comparative data using phylogenetically independent contrasts. Syst Biol 41, 18-32.
Holloway, R. L. 1979 Brain size, allometry, and reorganization: toward a synthesis. In Development and Evolution of Brain Size: Behavioral Implications (ed. M. E. Hahn, C. Jensen & B. C. Dudek), pp. 59-88. New York: Academic Press.
Howard, C. V. & Reed, M. G. 1998 Unbiased Stereology - Three-Dimensional Measurement in Microscopy. New York: Springer-Verlag.
Lavenex, P., Lavenex, P. B., Bennett, J. L. & Amaral, D. G. 2009 Postmortem changes in the neuroanatomical characteristics of the primate brain: hippocampal formation. J Comp Neurol 512, 27-51.
Mirra, S. S., Heyman, A., McKeel, D., Sumi, S. M., Crain, B. J., Brownlee, L. M., Vogel, F. S., Hughes, J. P., van Belle, G. & Berg, L. 1991 The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology 41, 479-86.
Mouton, P. R. 2002 Principles and Practices of Unbiased Stereology: An Introduction for Bioscientists. Baltimore and London: The Johns Hopkins University Press.
Paxinos, G., Huang, X. F. & Toga, A. W. 2000 The Rhesus Monkey Brain in Stereotaxic Coordinates. San Diego, CA: Academic Press.
Petrides, M., Cadoret, G. & Mackey, S. 2005 Orofacial somatomotor responses in the macaque monkey homologue of Broca's area. Nature 435, 1235-8.
Petrides, M. & Pandya, D. N. 1999 Dorsolateral prefrontal cortex: comparative cytoarchitectonic analysis in the human and the macaque brain and corticocortical connection patterns. Eur J Neurosci 11, 1011-36.
Rajkowska, G. & Goldman-Rakic, P. S. 1995 Cytoarchitectonic definition of prefrontal areas in the normal human cortex: I. Remapping of areas 9 and 46 using quantitative criteria. Cereb Cortex 5, 307-22.
Rilling, J. K. & Insel, T. R. 1999 The primate neocortex in comparative perspective using magnetic resonance imaging. Journal of Human Evolution 37, 191-223.
Rosabal, F. 1967 Cytoarchitecture of the frontal lobe of the squirrel monkey. J Comp Neurol 130, 87-108.