Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Disease-related microglia heterogeneity in the hippocampus of Alzheimer’s disease, dementia with Lewy bodies, and hippocampal sclerosis of aging

Disease-related microglia heterogeneity in the hippocampus of Alzheimer’s disease, dementia with... Introduction: Neuropathological, genetic, and biochemical studies have provided support for the hypothesis that microglia participate in Alzheimer’s disease (AD) pathogenesis. Despite the extensive characterization of AD microglia, there are still many unanswered questions, and little is known about microglial morphology in other common forms of age-related dementia: particularly, dementia with Lewy bodies (DLB) and hippocampal sclerosis of aging (HS-Aging). In addition, no prior studies have attempted to compare and contrast the microglia morphology in the hippocampus of various neurodegenerative conditions. Results: Here we studied cases with pathologically-confirmed AD (n =7), HS-Aging (n = 7), AD + HS-aging (n =4), DLB (n = 12), and normal (cognitively intact) controls (NC) (n = 9) from the University of Kentucky Alzheimer’s Disease Center autopsy cohort. We defined five microglia morphological phenotypes in the autopsy samples: ramified, hypertrophic, dystrophic, rod-shaped, and amoeboid. The Aperio ScanScope digital neuropathological tool was used along with two well-known microglial markers: IBA1 (a marker for both resting and activated microglia) and CD68 (a lysosomal marker in macrophages/microglia associated with phagocytic cells). Hippocampal staining analyses included studies of subregions within the hippocampal formation and nearby white matter. Using these tools and methods, we describe variation in microglial characteristics that show some degree of disease specificity, including, (1) increased microglia density and number in HS-aging and AD + HS-aging; (2) low microglia density in DLB; (3) increased number of dystrophic microglia in HS-aging; and (4) increased proportion of dystrophic to all microglia in DLB. Conclusions: We conclude that variations in morphologies among microglial cells, and cells of macrophage lineage, can help guide future work connecting neuroinflammatory mechanisms with specific neurodegenerative disease subtypes. Keywords: Aging, Microglia activation, Mixed dementia, Neurodegeneration, Neuroinflammation, Neuropathology * Correspondence: adam.bachstetter@uky.edu Sanders-Brown Center on Aging, University of Kentucky, 800 S. Limestone St, Lexington, KY, USA Full list of author information is available at the end of the article © 2015 Bachstetter et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 2 of 16 Introduction Materials and methods There is an increasing awareness that microglia may Human subjects have a pathogenic role in neurodegenerative diseases. Tissue samples that contained the hippocampus were ac- The discovery of genetic mutations in CD33 and quired from the UK-ADC biobank. Details of recruitment TREM2 associated with the risk of developing have been described previously [15]. Information including Alzheimer’s disease (AD) [1–4] has heightened the inter- demographic and neuropathologic data is presented est in defining microglia physiology and pathology in the (Table 1). The included cases (n = 39) represented a con- context of disease. Pio Del Rio-Hortega is credited with venience sample subdivided into groups as: NC, HS-aging, early insights into microglial pathology. He recognized AD, AD + HS-aging, or DLB. Cases represented approxi- that microglia are normally highly ramified and evenly mately age-matched sampling of the neuropathologically- distributed throughout the brain. He also noted that the defined diseases using the following criteria: AD (Braak > IV, morphology of microglia is dramatically altered in high density of neocortical amyloid plaques); isocortical response to central nervous system (CNS) pathology [5]. subtype of DLB; and HS-Aging (cell loss and gliosis out of As a molecular and functionally unique population of proportion to plaques/tangle pathology, with TDP-43 cells [6, 7], microglia exhibit a remarkable ability to pathology in the hippocampus). survey the brain and rapidly undergo a spectrum of responses to insults or tissue damage [8, 9]. The Immunostaining process by which microglia change shape, molecular Paraffin-embedded tissue sections were cut at 10-μm-thick. signature, and cellular physiology is defined as micro- Immunohistochemical (IHC) began with microwave anti- glia activation [5]. gen retrieval for 6 min (power 8) using Trilogy buffer (Cell The clinical diseaseformerlyreferredtosimplyas Marque; Rocklin, CA) for CD68 and Declere buffer (Cell “Alzheimer’sdisease” is, at the population level, a complex Marque; Rocklin, CA) for IBA1. Sections were then placed manifestation of many different brain conditions [10]. in 3% H O in methanol for 30 min. Following washes in 2 2 These age-related brain pathologies include AD (character- distilled water, sections were blocked in 5% goat serum at ized by amyloid plaques and neurofibrillary tangles), as well room temperature for 1 h. Sections were incubated in pri- as cerebrovascular disease, dementia with Lewy bodies mary antibodies IBA1 (rabbit polyclonal, 1:1,000 IHC, (DLB), and hippocampal sclerosis of aging (HS-Aging) [11]. Wako); CD68 (clone KP1) (1:50 IHC, Dako) overnight at Although each of these disorders seems to have a distinct 4°C. A biotinylated secondary antibody (Vector Laboratories) genetic, clinical, and pathological cluster of characteristics, was amplified using avidin-biotin substrate (ABC solution, to date there has not been characterization of the microglial Vector Laboratories catalog no. PK-6100), followed by color responses in these conditions. development in Nova Red (Vector Laboratories). Immuno- We sought to address questions related to microglial fluorescence (IF) staining was done following microwave morphology in neurodegenerative disease tissue: 1) Is antigen retrieval for 6 min (power 8) using Declere buffer microglia pathology seen only in the presence of amyloid (Cell Marque; Rocklin, CA) for primary antibodies to: IBA1 or tau pathology, or can it be seen in other age-related (rabbit polyclonal, 1:250 IF, Wako); and PHF-1 (1:500 IHC neurodegenerative diseases?; 2) Is there microglial re- and IF, a kind gift from Dr Peter Davies, Bronx, NY), and vi- gional heterogeneity in the hippocampus (for example, sualized using appropriate secondary antibody conjugated to gray matter only)?; and, 3) Can digital neuropathological an Alexafluor probe (1:200, Lifetechnologies) applied for 1 h. quantification detect differences in microglia activation A 0.1% solution of Sudan Black was used to reduce autofluo- in different neurodegenerative diseases? To address rescence. Slides were coverslipped using Vectashield mount- these questions, we queried well-characterized brain ing medium with DAPI (Vector Labs, Burlingame, CA). samples from the University of Kentucky Alzheimer’s Disease Center (UK-ADC) cohort. Specifically, brain tis- Quantitative image analysis sue was analyzed, incorporating multiple disease condi- Three different methods of quantitative image analysis tions, using two antibodies that react with microglia. were used in this study: 1) digital positive pixel algo- The CD68 antibody stains for a lysosomal-associated rithm, 2) digital nuclear algorithm, 3) and manual protein in macrophages/microglia and is associated with counting of IBA1 microglia (only in CA1 region). phagocytic cells [12, 13]. The IBA1 (ionized calcium Briefly, the Aperio ScanScope XT digital slidescanner binding adaptor molecule 1) antibody [14] is used widely was used to image the entire stained slide at 40x magni- as a pan marker for both resting and activated microglia. fication to create a single high-resolution digital image. Using these two widely studied microglia markers, CD68 The Aperio positive pixel count algorithm (version 9) and IBA1, we defined microglia morphologies in the was used to quantify the amount of specific staining in aged brain, including some features that show evidence the region, and the Aperio nuclear algorithm (version 9) of disease specificity. was used to determine the number of stained microglia Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 3 of 16 Table 1 Cohort demographics and numbers age at Final ApoE PMI Braak CERAD plaque diffuse neuritic diffuse Lewy case sex NFTs death MMSE alleles (h) stage stage plaques plaques bodies NC = non-demented control: mean age =86; mean MMSE = 30; Median Braak stage = 2; Median CERAD = 0 1 81 30 M 3 / 3 2.17 2 0 0 0 0 0 2 91 30 M n/a n/a 2 0 0 0 1.4 0 3 86 30 M 3 / 3 2.17 3 0 0 0 4.6 0 4 93 30 F 3 / 3 2.25 2 0 0 0 0.25 0 5 84 30 M 2 / 3 3.25 0 0 n/a n/a n/a 0 6 85 30 M 3 / 3 2 3 1 2 0.5 8 0 7 84 30 F 3 / 3 2.42 0 0 0 0 0 0 8 92 30 F 2 / 3 3.25 3 2 5 0 0.75 0 9 81 30 M 3 / 4 2 2 2 2 10.5 1 0 HS = hippocampal sclerosis of aging: mean age =87; mean MMSE = 22.7; Median Braak stage = 2; Median CERAD = 0 10 74 20 M 3 / 4 8 3 2 0 0 1.4 0 11 95 16 F 3 / 4 3.25 3 0 0 0 1 0 12 87 28 F 3 / 3 1.82 3 0 0 0 4 0 13 84 10 F n/a 2.57 2 0 0 0 0 0 14 91 29 F 2 / 3 2.87 2 1 0 0 6.8 0 15 91 30 M 3 / 3 2.83 2 0 n/a n/a n/a 0 16 88 26 M 3 / 4 2.33 0 0 0 0 0 0 AD = Alzheimer’s disease: mean age =77; mean MMSE = 11; Median Braak stage = 6; Median CERAD = 3 17 75 18 F 3 / 4 2.5 6 3 9 0.33 25 0 18 84 13 M 4 / 4 5.17 6 3 4 1.33 26.75 0 19 65 3 F 3 / 4 4.1 6 3 2.67 1.67 41 0 20 85 4 F 3 / 3 11.2 6 3 6.67 1.33 78.8 0 21 79 6 M n/a 2.08 6 3 4.33 2.33 25.6 0 22 67 11 M 2 / 3 1.75 6 3 0.33 0 2.4 + 23 82 25 M n/a 2.75 6 2 1.67 2.67 19.8 + AD + HS: mean age =91; mean MMSE = 7.8; Median Braak stage = 6; Median CERAD = 3 24 96 6 F 3 / 3 6.75 6 3 n/a n/a n/a 0 25 91 13 F 3 / 3 3 5 3 6.67 1.33 34 0 26 91 12 F 4 / 4 2.33 6 3 0 0 10.25 + 27 87 0 F 3 / 3 2.67 6 3 0 0 54.6 0 DLB = Dementia with Lewy bodies: mean age =80; mean MMSE = 17.25; Median Braak stage = 2; Median CERAD = 1 28 65 9 M 4 / 4 9.5 2 3 2 7.67 2 + 29 61 18 M 3 / 3 2 2 0 0 0 9.6 + 30 85 2 F 2 / 3 2 2 3 4 6.67 3.6 + 31 85 27 M 3 / 3 11.2 2 1 0 0 0.8 + 32 89 27 M 3 / 3 2.42 2 0 0 0 8.5 + 33 92 27 F 3 / 4 2.42 2 1 0.67 0 2 + 34 68 11 M 3 / 4 3.75 2 2 0 0 0.4 + 35 81 15 M 3 / 3 2.42 2 2 1.5 2.5 0.25 + 36 78 9 M 3 / 3 2.5 1 0 0 0 2 + 37 81 26 M 3 / 3 5.77 1 1 0 0 0 + 38 97 21 M 2 / 3 3.5 1 0 0 0 1.8 + Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 4 of 16 Table 1 Cohort demographics and numbers (Continued) 39 78 15 M 3 / 4 n/a 1 1 3.5 0.5 0 + Counts in the hippocampus of neurofibrillary tangles (NFTs), and amyloid plaques, without neurites (diffuse plaques), and with degenerating neurites (neuritic plaques) (see [42]). Abbreviations: Apo E apolipoprotein E; PMI = post-mortem interval; MMSE = mini-mental state examination; +, feature present; n/a = not available as previously described [16, 17]. The number of IBA1 Statistics microglia was counted by morphological appearance in JMP Software version 10.0 was used for statistical analysis. 5 arbitrarily placed 250 × 250 μmboxes in theCA1 re- Normality was assessed using the Shapiro-Wilk test. As gion. A researcher (coauthor ADB) blind to all samples’ there were only a few violations of normality, and ANOVA case histories conducted all data analysis. Immunofluor- is robust to such violations [18], a one-way ANOVA escence was imaged using a Nikon Eclipse 90i upright followed by a Tukey post hoc analysis was used to compare microscope equipped with a Nikon DS-Ri1 digital differences between the five groups. Mean ± SD for quanti- camera. fications are shown in Table 2. Differences between means Table 2 Summary of microglia neuropathological assessment CD68 positive pixels (Fig. 2) WM sub CA1 CA2/3 CA4 DG hipp ave NC (N = 9) 7.5 ± 4 2.5 ± 1.6 2.6 ± 2.2 3.2 ± 2.1 2.8 ± 2.2 2.4 ± 1.6 3.5 ± 2.1 HS-aging (N = 7) 8.7 ± 2.4 6 ± 3.5 5.4 ± 2.1 4.1 ± 2.1 3.4 ± 1.2 2.4 ± 1.2 5 ± 1.4 AD (N = 7) 10.6 ± 3.2 6.2 ± 2 5.3 ± 1.5 4.2 ± 2.6 3.5 ± 1.2 4.8 ± 2 5.8 ± 1 AD + HS (N = 4) 8.7 ± 3.6 5.5 ± 2.6 5.8 ± 2.1 3.2 ± 1.2 2.4 ± 1.3 2.5 ± 1.4 4.7 ± 1.2 DLB (N = 12) 7.6 ± 2.8 2.7 ± 1 2.3 ± 0.8 3.4 ± 2.3 2.7 ± 1.1 2.2 ± 0.3 3.5 ± 0.9 CD68 nuclear algorithm (Fig. 4) NC (N = 9) 54.7 ± 52.1 13.9 ± 17.9 16 ± 18.5 17 ± 18.3 20 ± 24 10.7 ± 11.3 22.1 ± 22.1 HS-aging (N = 7) 54.8 ± 45.1 34.4 ± 33.6 30.2 ± 20.9 20.4 ± 19.8 19.2 ± 11.9 8.1 ± 7.8 27.8 ± 21.2 AD (N = 7) 100.2 ± 68.4 45.8 ± 35.8 29.2 ± 16.8 16.4 ± 7.5 17.3 ± 13.3 26.3 ± 18.1 39.2 ± 22.4 AD + HS (N = 4) 32.4 ± 15.8 27.2 ± 22.4 28.7 ± 26.6 15.4 ± 14.4 8 ± 6.8 3.8 ± 5.2 19.2 ± 13.7 DLB (N = 12) 53.3 ± 35 11.8 ± 9 11.6 ± 7.7 15.6 ± 11.1 13.3 ± 8.1 8.9 ± 5.9 19.1 ± 10.1 IBA1 nuclear algorithm (Fig. 5) NC (N = 9) 101.9 ± 30.2 66.5 ± 14.3 60.8 ± 22.9 74 ± 11.7 82.7 ± 16.8 80.5 ± 18.7 77.7 ± 11.7 HS-aging (N = 7) 143.7 ± 80.8 84.7 ± 39.4 122.1 ± 63.5 142.8 ± 37.3 110.9 ± 50.3 92.3 ± 39.1 116.1 ± 45.4 AD (N = 7) 87.8 ± 55.4 84.1 ± 44.2 64.2 ± 16.2 75.5 ± 19.2 43.2 ± 27.2 50.5 ± 37.2 67.5 ± 26.7 AD + HS (N = 4) 97.9 ± 25.8 116.4 ± 74.9 108.2 ± 42.1 132.4 ± 64.6 94.3 ± 47.1 98.9 ± 47.1 108 ± 41.9 DLB (N = 12) 103.2 ± 55.6 69.3 ± 35.6 70 ± 32 74.2 ± 46.7 83.8 ± 48 74.5 ± 43.9 79.2 ± 40.7 IBA1 positive pixels (Fig. 6) NC (N = 9) 2.8 ± 0.8 2.6 ± 0.9 2.4 ± 0.7 3.1 ± 0.8 3.3 ± 1 3.1 ± 0.8 2.9 ± 0.7 HS-aging (N = 7) 4 ± 1.8 2.5 ± 0.9 3.2 ± 1.1 3.9 ± 0.8 3.5 ± 1.3 3.1 ± 1.1 3.3 ± 1 AD (N = 7) 2.7 ± 1.2 2.5 ± 1.1 2.2 ± 0.6 3.1 ± 1 2 ± 1.1 2.1 ± 0.8 2.4 ± 0.7 AD + HS (N = 4) 3.6 ± 0.8 3.4 ± 2.1 3.6 ± 1.1 3.9 ± 1.2 3.1 ± 0.7 3.3 ± 1 3.5 ± 1 DLB (N = 12) 2.8 ± 0.8 1.9 ± 0.6 1.9 ± 0.5 2.3 ± 0.8 2.6 ± 0.9 2.5 ± 0.9 2.3 ± 0.7 Morphological assessment of IBA1 microglia in CA1 region (Fig. 7) ramified hypertrophic dystrophic rod-shaped amoeboid total NC (N = 9) 16.9 ± 9.8 3.7 ± 7.1 4.4 ± 3.9 2.1 ± 4.1 3.7 ± 5 30.8 ± 10.3 HS-aging (N = 7) 2.6 ± 2.6 13.8 ± 11.9 32.2 ± 22 2.4 ± 5.1 8.8 ± 8.8 59.9 ± 29.8 AD (N = 7) 19.7 ± 10.4 4.5 ± 6.1 11.2 ± 9.7 2.6 ± 2.7 4.5 ± 3.6 42.4 ± 11.5 AD + HS (N = 4) 9.8 ± 8.3 20.2 ± 13.7 19.8 ± 11.2 2.6 ± 3.2 14.6 ± 12.4 67 ± 15.8 DLB (N = 12) 9.3 ± 6.5 2.5 ± 3.6 13.9 ± 9.3 3.9 ± 4 1.9 ± 2.6 31.3 ± 8.2 Values represent mean ± SD for the quantification of CD68 and IBA1 immunohistochemistry. Data is plotted in the indicated figures Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 5 of 16 were considered significant at p < 0.05. Heatmaps were gen- of the cerebral neocortex. In pure DLB, there are low levels erated using JMP Software version 10.0. All other graphs of amyloid-β pathology or NFTs, as shown in Table 1. were generated using GraphPad Prism software version 6.0, Primary goals of this study were to assess regional with values expressed as mean ± SEM. microglia heterogeneity and to exploit the ability of digital neuropathological quantification to detect in differences microglial morphometry when cases are stratified accord- Results ing to their neurodegenerative diseases. Six regions of Five groups of cases (Table 1) were pathologically-confirmed interest (ROI) were identified by dividing the hippocampal as either AD (n=7), HS-Aging (n =7), AD +HS-aging (n = formation into the dentate gyrus (DG), the cornu ammo- 4), DLB (n = 12), and NC (n = 9). HS-aging and DLB cases nis (CA) areas (CA1, CA2/3, and CA4), the subiculum were included in this study to determine if there is disease (sub), and the adjacent white matter (WM) (Fig. 1). Repre- specificity in microglia pathology and to provide the first sentative examples of the ROIs are shown in Fig. 1. quantitative analysis of microglia in HS-Aging. Pure HS-aging cases lacked substantial additional patholo- Pattern of CD68 staining in the hippocampus of autopsy gies AD-type pathology, or Lewy bodies [19–21], as cases shown in Table 1. The neuropathological changes asso- Quantification of the CD68 positive pixels is shown in ciated with neocortical/diffuse Lewy body disease in- Fig. 2. By a one-way ANOVA a significant effect of dis- clude, by definition, α-synuclein immunoreactive neuronal ease status was found sub (Fig. 2c; F = 6.3001; p = 4,38 inclusions (Lewy bodies) and processes in multiple portions 0.0007), CA1 (Fig. 2d; F = 8.0944; p < 0.001), DG 4,38 Fig. 1 Regions of interest used for microglia analysis. A representative hippocampus is shown for the five neuropathological diagnoses included in this study. The outlines illustrate the boundaries used in identifying the following brain regions: white matter (WM), subiculum (sub), the cornu ammonis(CA)areas,CA1,CA2/3,CA4, and thedentate gyrus(DG). TheROIsshowninthe figure arenot theactualROIsusedfor analysis,assomeofthe ROIs (WM and sub) could not be included in the image frame, as the brain region was larger than the image frame Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 6 of 16 Fig. 2 Digital neuropathological quantification using positive pixel algorithm of CD68 immunostaining in the hippocampus of autopsy cases. Representative example of (a) CD68 staining and a digitally generated mark-up showing the ability of the positive pixel algorithm to detect the staining. Digital neuropathological quantification of the CD68 staining using the positive pixel algorithm is shown for the (b) WM, (c) sub, (d) CA1, (e) CA2/3, (f) CA4, (g) DG, and for the (h) average of the hippocampal formation. Circles represent an individual case, with mean and SEM shown § ‡ for the group. Statistical comparisons: *p <0.05 compared to AD cases. p<0.05comparedtoHS-agingcases. p < 0.05 compared to AD + HS-aging cases. (i) Heatmap summarizes the results shown in (b-h) (also see Table 2) (Fig. 2g; F = 5.3332; p = 0.0019), and in the average regional and disease-specific heterogeneity in the stain- 4,38 of the six regions in the hippocampus formation ing (Fig. 3). Of note is a large round cell type that can (Fig. 2h; F = 4.3221; p = 0.0062). No significant effect be found in areas of high density staining as shown in 4,38 was found by a one-way ANOVA in WM (Fig. 2b), Fig. 3b-c. Interestingly, just distal to the very intense + + CA2/3, (Fig.2e),orCA4 (Fig.2f).HS-aging, AD,and accumulation of CD68 cells, the CD68 staining was AD + HS-aging were found to have significantly more unremarkable, with a few ramified microglia (Fig. 3d). + + CD68 staining in the CA1 region compared to NC or Quantification of the number of large round CD68 DLB cases (Fig.2d).However,there was nosignificant cells was done using the nuclear algorithm, by adjusting differenceamong thethree disease conditions(HS- the algorithm to detect only the large round cells as aging, AD, and AD + HS-aging) in the CA1 region shown inFig.4.Incomparisontodesignbased stereo- (Fig. 2d). Interestingly, we found significantly more logical methods, limitations of the nuclear algorithm CD68 staining in the DG of AD cases compared to the include an inability to provide an estimate of the total other four groups (Fig. 2g). When averaged across the number of microglia, because of a lack of 3- six-hippocampal formation sub regions, the AD cases dimensional volume measurements [22, 23]. Limita- were found to have significantly more CD68 staining tions notwithstanding, results of the nuclear algorithm compared to NC or DLB groups. Overall, the greatest were similar to the positive pixel algorithm, with the CD68 staining was seen in the WM, as is evident by HS-aging and AD groups having the greatest number the heatmap summary of the CD68 positive pixel of CD68 cells (Table 2). As shown by the heatmap, the + + analysis (Fig. 2i). A survey of the CD68 staining in the greatest number of CD68 cells was found in the WM six-hippocampal formation regions illustrates the of AD cases (Fig. 4). Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 7 of 16 + + Fig. 3 Survey of CD68 staining in the hippocampus of autopsy cases. (a) Representative examples of CD68 staining pattern in the brain regions analyzed by digital neuropathological analysis. (b) Low power photomicrograph of hippocampus of a DLB individual (case #36) highlights an area of intense staining (blue arrow) shown in (c), and an area of low CD68 staining (black arrow) in a nearby region (d) Digital quantification of IBA1 staining in the of disease status was found in the CA1 region (Fig. 5d; hippocampus of autopsy cases F = 3.9914; p = 0.0092), CA2/3 region (Fig. 5e; F = 4,38 4,38 Quantification of the number of IBA1 cells by the nuclear 5.8525; p = 0.0011), and in the CA4 (Figure 5f F = 4,38 algorithm is shown in Fig. 5. A representative example of 2.6929; p = 0.0473). No significant effect was found by a the ability of the algorithm to detect individual cells is one-way ANOVA in WM (Figure 5b), sub (Fig. 5c), DG shown in Fig. 5a. By a one-way ANOVA, a significant effect (Fig. 5g), or in the average of the six regions in the Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 8 of 16 Fig. 4 Digital neuropathological quantification using nuclear algorithm of number of large round CD68 cells in the hippocampus of autopsy cases. Representative example of (a) CD68 staining and a digitally generated mark-up showing the ability of the nuclear algorithm to detect the staining of the large round cells, but not smaller cells or processes. Digital neuropathological quantification of the CD68 staining using the nuclear algorithm to detect the staining is shown for the (b)WM, (c)sub (F = 2.9934; p = 0.0321), (d)CA1,(e)CA2/3,(f)CA4,(g)DG (F = 4.3393; p =0.0061), 4,38 4,38 and for the (h) average of the hippocampal formation. Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: *p <0.05 compared to AD cases. (i) Heatmap summarizes the results shown in (b-h) (also see Table 2) hippocampus formation (Fig. 5h). In the CA1 region, the microglia density and cell number were measured irrespect- HS-aging had an increased number of IBA1 cells com- ive of the microglia morphology. For example, a striking pared to NC, AD or DLB. As shown by the heatmap sum- pattern of IBA1 microglia morphology is the rod-shaped mary, a similar pattern of increased number of IBA1 microglia, which were readily apparent in a subset of cases. microglia was found in the HS-aging and AD + HS-aging As shown in Fig. 7b-c, rod-shaped microglia are character- groups compared to the NC, AD, or DLB groups ized by a narrow cell body with a few planar processes. The (Fig. 5i). Quantification of the IBA1 positive pixels (Table 2) rod-shaped microglia could be found as individual cells also showed a similar pattern of increased IBA1 staining (Fig. 7b), or as long and thin groups of cells that may have in the HS-aging and AD + HS-aging groups compared to fused(Fig.7b and c andFig.8).Theappearanceofmicro- the NC, AD, or DLB groups (Fig. 6). glia with polarized and parallel processes suggested that the microglia could be following neurites—possibly, degenerat- IBA1 microglia morphology in the hippocampus of ing axons or neurons themselves. To test the possibility that autopsy cases microglia could be surrounding degenerating neuronal pro- An examination (Fig. 7a) of the IBA1 microglia in the six cesses, double label immunofluorescence was performed ROIs in the five neuropathologic groups showed remark- for microglia (IBA1) and NFTs (PHF1). Fig. 8a shows abun- + + able heterogeneity in microglia density, as captured by the dant PHF1 staining and IBA1 rod-shaped microglia in digital neuropathological quantification. There was also het- theCA1 region of an AD individual (case#23).Wefound + + erogeneity in IBA1 microglia morphology, which was un- no evidence of systematic overlap of PHF1 neurites and derappreciated in the digital neuropathological analysis, as IBA1 rod-shaped microglia, as shown in Fig. 8b. Rather, Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 9 of 16 Fig. 5 Digital neuropathological quantification using nuclear algorithm of number of IBA1 cells in the hippocampus of autopsy cases. Representative example of (a) IBA1 staining and a digitally generated mark-up showing the ability of the nuclear algorithm to detect six stained cells. Digital neuropathological quantification of the IBA1 staining using the nuclear algorithm is shown for the (b) WM, (c) sub, (d) CA1, (e) CA2/ 3, (f) CA4, (g) DG, and for the (h) average of the hippocampal formation.. Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: p < 0.05 compared to HS-aging cases. (i) Heatmap summarizes the results shown in (b-g) (also see Table 2) long trains of rod-shaped microglia could sometimes be (Fig. 10a), to allow measurement of changes in the micro- seen to run parallel to and between PHF1 neurons but did glia classes associated with the five neurodegenerative dis- not co-localize with the PHF1 staining (Fig. 8c). In this ex- ease groups. The five classes of microglia morphologies ample, the tip of the rod-shaped microglia was near (but included: 1) ramified microglia, which have a ‘surveying’ not within) a PHF1 structure, and the IBA-1 immunoreac- non-reactive microglia morphological appearance, with thin tive structure appeared to be a fusion / cluster of multiple highly branched processes [5, 26]; 2) hypertrophic microglia cells with 5 clearly visible DAPI nuclei (Fig. 8d). (often called activated microglia), which have become en- Another pattern of microglia morphology observed larged, hyper-ramified or may have short thick processes [5, was the dystrophic / degenerating microglia, which over- 26]; 3) dystrophic microglia, with processes that are spher- lapped morphologically with cells that have been de- oidal, beaded, de-ramified, or fragmented [24–26]; 4) rod- scribed to have processes that are spheroidal, beaded, shaped microglia, characterized by a narrow cell body with de-ramified, or fragmented [24, 25]. Examples of dys- a few planar processes [27, 26]; and 5) amoeboid microglia, trophic / degenerating microglia are shown in Fig. 9. In with an enlarged cell body with few to no processes [5, 26]. AD (Fig. 9a) and DLB (Fig. 9b), for example, the dys- CD68 staining could clearly identify cells with an amoeboid trophic / degenerating microglia had very thin processes morphology, and to a lesser extent cells with a ramified that are beaded and fragmented. In HS-aging (Fig. 9c) morphology. In contrast, IBA1 staining was useful to iden- and AD + HS-aging (Fig. 9d), dystrophic microglia tify all five microglia morphologies. Therefore, IBA1 stain- morphology was more striking, and the processes of the ing was used to quantify the distribution in the microglia microglia were beaded and tortuous. morphology according to these five subtypes of microglial The remarkable diversity in the microglia morphology shapes. Focusing on the CA1 region of the hippocampus, led us to carefully review and categorize the morphological the numberofeachofthe five morphological classesof appearances of the microglia into five distinct classes IBA1 microglia was counted in five randomly placed Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 10 of 16 Fig. 6 Digital neuropathological quantification using positive pixel algorithm of IBA1 immunostaining in the hippocampus of autopsy cases. Representative example of (a) IBA1 staining and a digitally generated mark-up showing the ability of the positive pixel algorithm to detect the staining. Digital neuropathological quantification of the IBA1 staining using the positive pixel algorithm is shown for the (b) WM, (c) sub, (d) CA1 (F =5.0943; p =0.0025), (e)CA2/3 (F =4.8888; p = 0.0032), (f)CA4, (g) DG,, and for the (h) average of the hippocampal formation (F =3.0201; 4,38 4,38 4,38 p = 0.0311). Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: *p <0.05 compared to § ‡ AD cases. p < 0.05 compared to HS-aging cases. p < 0.05 compared to AD + HS-aging cases. (i) Heatmap summarizes the results in (b-h) (also see Table 2) and evenly distributed 250x250μm regions of interest (p = 0.0035) cases. HS-aging cases had more total (ROI). HS-aging cases had fewer ramified microglia microglia than NC (p = 0.0072) or DLB (p = 0.0048) than NC (p = 0.0091) or AD (p = 0.0027) cases (Fig. 10b, cases (Fig. 10g, Table 2). As the total number of micro- Table 2). HS-aging and AD + HS-aging had the most hyper- glia was found to be altered in the different groups, trophic microglia. AD + HS-aging cases had more hyper- each of the five microglia classifications was plotted as trophic microglia than NC (p = 0.0132), AD (p = 0.0270), or a percentage of the total number of microglia (Fig. 10h) DLB (p = 0.0044) cases, and HS-aging cases had more to help visualize the microglia morphology distribu- hypertrophic microglia than DLB (p = 0.0140) cases tions within and among the different diseases. (Fig. 10c, Table 2). HS-aging cases had more dystrophic microglia than NC (p = 0.0005), AD (p = 0.0193), or Discussion DLB (p = 0.0225) cases (Fig. 10d, Table 2). Quantification of The present study underscores the rich diversity of micro- rod-shaped microglia identified a subset of cases with glial morphologies in the hippocampus of the human brain abundant rod-shaped microglia; however, the cases that maychangeaccording to thediseasesofaging.Weob- were not specific to a disease group (Fig. 10e, Table 2). served regional heterogeneity in the hippocampal formation + + AD + HS-aging cases had more amoeboid microglia in the density and number of IBA1 and CD68 microglia. than NC (p = 0.0428), or DLB (p = 0.0085) cases We also observed five morphologically-defined classes of (Fig. 10f, Table 2). The total number of microglia in the IBA1 labeled microglia: ramified, hypertrophic, dystrophic, CA1 region, regardless of morphology, was greatest in rod-shaped, and amoeboid (Fig. 10). Our observations pro- HS-aging and AD + HS-aging. AD + HS-aging cases vide evidence for subclasses of microglial morphologies that had more total microglia than NC (p = 0.046), or DLB are seen in particular neurodegenerative diseases. The data Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 11 of 16 + + Fig. 7 Survey of IBA1 staining in the hippocampus of autopsy cases. (a) Representative examples of IBA1 staining pattern in the brain regions analyzed by digital neuropathological analysis(b) A low powered photomicrograph shows the widespread distribution of rod shaped microglia in the CA1 region of a DLB individual (case #34). Long trains of microglia (highlighted by blue arrows) are shown at higher magnification in (c). provide at least some support for disease-specific microglia neuropathological quantification to measure changes in pathology in age-related dementias. human microglia activation and compare directly the A primary goal of our project was to determine if microglia response in the different neurodegenerative digital neuropathological quantification could detect diseases, and the first to assess microglia in HS-aging disease-specific changes in IBA1 and CD68 labeled cases. The digital neuropathological quantification was microglia activation. This is the first study to use digital able to detect regional differences in IBA1 and CD68 Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 12 of 16 + + Fig. 8 Lack of localization of IBA rod-shaped microglia to PHF1 neurons in an AD individual (case #23). (a) A low powered photomicrograph shows the distribution of rod-shaped microglia next to PHF1 cells. (b) A linear group of rod-shaped microglia is shown at a higher magnification. (c) A second example of rod-microglia, where the microglia run parallel and between PHF1+ neurons. (d)Ofnote, thepolar endof the rod-microglia(white arrow)was foundtohave5DAPI nuclei staining associated with the neuropathological diagno- cases, suggesting that the HS-aging pattern of microglia sis. Specifically, we found increased IBA1 and CD68 staining is dominant over the AD pattern, and that staining in the HS-aging and AD + HS-aging cases. there is not a robust additive effect of the two patholo- Interestingly, the spatial pattern and magnitude of the gies. Thus, results of the digital neuropathological changes in IBA1 and CD68 staining were remarkably quantification clearly show a pattern of microglia acti- similar between the HS-aging and AD + HS-aging vation associated with a specific neurodegenerative Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 13 of 16 + + Fig. 9 Dystrophic IBA1 microglia in the hippocampus. Examples of IBA1 dystrophic microglia in the CA1 region of AD individual (a; case #20), DLB individual (b; case #33), HS-aging individual (c; case #15), and AD + HS-aging individual (d; case #27). Scale bar is 25 μm disease, but overall the quantification provided only of function; see [25, 28]). Support for the hypofunctional modest sensitivity, with limited diagnostic potential, to (as opposed to activated) microglia model is largely based separate AD from HS-aging. To determine the repro- on morphological examination of IBA1-stained microglia in ducibility of the digital quantification, 12 of 39 cases se- autopsy samples from aged humans [29–32], as currently lected at random were replicated in an independent there are no specific markers that recognize only degenerat- experiment. Even with the modest sample size, a com- ing/dystrophic microglia. In addition, the dystrophic micro- parison of the number of IBA1 positive pixels in the glia phenotype seen in humans is largely absent in rodent CA1 regions between the two independent experiments models [25]. This may reflect intrinsic differences in human resulted in a R = 0.7. The results of the replication microglia [33], or may reflect limitations in the current ani- study support the use of digital neuropathological quan- mal models. We found that aged individuals without de- tification as a relatively accurate, unbiased, quantitative and mentia were more likely to have ramified microglia than efficient means of neuropathological assessment. In the fu- individuals with dementia (AD, HS-aging, AD + HS-aging, ture, development of algorithms that can detect the differ- or DLB). Moreover, the present study confirmed that dys- ent microglia phenotypes (Fig.10) should greatlyimprove trophic microglia are found in aged individuals and in in- the potential of this approach to detect disease-related creased numbers in aged individuals with three distinct changes in microglia morphology, until specific molecular forms of dementia (AD, HS-aging, and DLB). Our results markers that recognize the different microglia morpho- provide an independent confirmation of the presence of logical states are available. dystrophic microglia described by the Streit laboratory The long-standing view that microglia become activated [29–32]. Research at our center has previously shown and promote neuroinflammation in neurodegenerative dis- differences in the M1/M2 microglia phenotype between ease (toxic gain of function) has been challenged recently mild AD and end-stage AD [34], supporting changes in by the concept of the dystrophic / diseased microglia (loss the temporal dynamic of the microglia response to Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 14 of 16 Fig. 10 Disease specific patterns in IBA1 microglia morphology. (a) Representation of microglia morphologies seen in the hippocampus of aged individuals. The number of microglia was quantified at 40x magnification in five 250 x 250 μm regions of interest (ROIs) that were randomly placed and evenly spaced in the CA1 region. Following the classification shown in (a), IBA microglia were classified as either (b) ramified (F = 5.3533; p = 0.0019), 4,38 (c)hypertrophic(F = 5.5082; p = 0.0016), (d) dystrophic (F = 5.7249; p = 0.0012), (e) rod-shaped, or (f) amoeboid (F = 3.9836; p = 0.0093). (g)The 4,38 4,38 4,38 number of microglia (F = 7.2694; p = 0.0002) in the five classifications was summed to get the total number of microglia. The gray circles in 4,38 (b-g) represent the average number of microglia per mm for an individual case, with mean and SEM shown for each group (see also Table 2). § ‡ Statistical comparisons: p < 0.05 comparedtoHS-agingcases. p < 0.05 compared to AD + HS-aging cases. (h)Asthe totalnumberof microglia significantly varied by group, the number of microglia in each of the five classifications was plotted as a percent of the total number of microglia to illustrate the disease-related patterns in microglia morphology (also see Table 2) Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 15 of 16 varying degrees of neuropathology. Still, the temporal able to provide average cell body size and roundness, along dynamic of microglia in humans over the course of the with the number of processes, process length and volume lifespan has not been defined and is not fully testable occupied by the processes [40].That study didnot include through autopsy (cross-sectional) studies. This is a vital any samples without neurologic disease, and therefore area for future investigation. underestimated the heterogeneity in microglia morphology; The current study highlights the importance of for example, they did not describe rod-shaped microglia. morphology-based readout of cell activity. Rod-shaped Using a similar approach, others have attempted to define microglia are a particularly fascinating microglia phenotype, classes of microglia morphology, such as the rod-shaped which was first described by Nissl more than 100 years ago microglia, by calculating celllengthto cellwidth andthe (reviewed in: [27]). Rod-shaped microglia have been de- number of polar vs. planar branches [36]. Moreover, others scribed clinically in neurosyphilis, subacute sclerosing have proposed digital 3D reconstruction of the microglia as panencephalitis, lead encephalopathy, viral encephalitis a means to quantify the microglia morphology [41]. How- including HIV-1, and Rasmussen's encephalitis [27, 35]; ever, before microglia morphological assessment can be- however, there are few modern reports of rod-shaped come standard practice in characterizing the microglia microglia in the clinical literature. In experimental pathology, a consensus must be established on what defines models, rod-shaped microglia have been best described different microglia morphologies, as there is currently no following traumatic brain injury [36–38, 16], where a consensus-based agreement on definitions, or terminology diffuse braininjurywill causethe rapid (by 6 h) for the specific classes of microglia morphology. Our study polarization of microglia to follow along neuronal pro- provides a first step towards this goal and will hopefully cesses. It has been shown previously in rats that micro- provide a framework to move the field forward in this glia will fuse specifically to the apical dendrite of direction. neurons infected with a retrovirus, but not to un- Conflict of interests infected neurons [39]. It is not clear if fusion is occur- The authors declare that they have no competing interests ring in the case of rod-shaped microglia in the current Acknowledgements study. Beyond these few reported observations, little is We are profoundly grateful to all of the study participants who make this known mechanistically about the chemoattractant sig- research possible. The corresponding author, Adam Bachstetter, PhD, had full nals that drive formation of rod-shaped microglia, or access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Research reported in this about the specific functions of the rod-shaped micro- publication was supported by National Institutes of Health under award glia in relation to the neuron. We found that rod- numbers P30 AG028383, K99 AG044445. The content is solely the shaped microglia could be proximal and parallel to responsibility of the authors and does not represent the official views of the National Institutes of Health. PHF1 neurons/axons, but the rod-shaped microglia did not appear to fuse with or engulf the PHF1 struc- Author details tures. Rod-shaped microglia were present in approxi- Sanders-Brown Center on Aging, University of Kentucky, 800 S. Limestone St, Lexington, KY, USA. Department of Anatomy and Neurobiology, mately 60% of cases included in this study, but were University of Kentucky, Lexington, KY, USA. Department of Neurology, most abundant in a subset of cases. Review of the case University of Kentucky, Lexington, KY, USA. Department of Pathology and histories of individuals with abundant rod-shaped Laboratory Medicine, Division of Neuropathology, University of Kentucky, Lexington, KY, USA. Department of Epidemiology, University of Kentucky, microglia did not identify any obvious commonalities. Lexington, KY, USA. Department of Biostatistics, University of Kentucky, A goal for future studies will be to identify a larger Lexington, KY, USA. Department of Statistics, University of Kentucky, sample of cases with abundant rod-shaped microglia to Lexington, KY, USA. distinguish clinical-pathological correlations, as a first Received: 1 May 2015 Accepted: 4 May 2015 step in defining mechanistically the functions of this mysterious cell type. References A limitation of this study is the subjective criteria that 1. Malik M, Simpson JF, Parikh I, Wilfred BR, Fardo DW, Nelson PT, Estus S (2013) were used to classify microglia into one of the five morpho- CD33 Alzheimer's risk-altering polymorphism, CD33 expression, and exon 2 logical categories (Fig. 10). This approach requires an ex- splicing. J Neurosci 33(33):13320–13325. doi:10.1523/JNEUROSCI.1224-13.2013 2. Griciuc A, Serrano-Pozo A, Parrado AR, Lesinski AN, Asselin CN, Mullin K, perimenter capable of discerning differences in microglia Hooli B, Choi SH, Hyman BT, Tanzi RE (2013) Alzheimer's disease risk gene morphology. The approach also imparts bias and potential CD33 inhibits microglial uptake of amyloid beta. Neuron 78(4):631–643. for experimenter error. The current study demonstrates doi:10.1016/j.neuron.2013.04.014 3. Jonsson T, Stefansson H, Steinberg S, Jonsdottir I, Jonsson PV, Snaedal J, that there is a great diversity in the microglia morphology Bjornsson S, Huttenlocher J, Levey AI, Lah JJ, Rujescu D, Hampel H, Giegling in humans, which is underappreciated, as this diversity is I, Andreassen OA, Engedal K, Ulstein I, Djurovic S, Ibrahim-Verbaas C, Hofman A, largely absent from animal models. There have been prior Ikram MA, van Duijn CM, Thorsteinsdottir U, Kong A, Stefansson K (2013) Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med attempts to operationalize morphological changes in micro- 368(2):107–116. doi:10.1056/NEJMoa1211103 glia. A recent study reconstructed microglia from mice and 4. Guerreiro R, Wojtas A, Bras J, Carrasquillo M, Rogaeva E, Majounie E, humans using computer-based tracing systems, and was Cruchaga C, Sassi C, Kauwe JS, Younkin S, Hazrati L, Collinge J, Pocock J, Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 16 of 16 Lashley T, Williams J, Lambert JC, Amouyel P, Goate A, Rademakers R, 21. Dickson DW, Davies P, Bevona C, Vanhoeven KH, Factor SM, Grober E, Morgan K, Powell J, St George-Hyslop P, Singleton A, Hardy J, Alzheimer Aronson MK, Crystal HA (1994) Hippocampal sclerosis - a common Genetic Analysis G (2013) TREM2 variants in Alzheimer's disease. N Engl pathological feature of dementia in very old (greater-than-or-equal-to-80 years of J Med 368(2):117–127. doi:10.1056/NEJMoa1211851 age) humans. Acta Neuropathol 88(3):212–221 5. Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (2011) Physiology of 22. West MJ (1993) New stereological methods for counting neurons. Neurobiol microglia. Physiol Rev 91(2):461–553. doi:10.1152/physrev.00011.2010 Aging 14(4):275–285 6. Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, 23. Sterio DC (1984) The unbiased estimation of number and sizes of arbitrary Koeglsperger T, Dake B, Wu PM, Doykan CE, Fanek Z, Liu L, Chen Z, particles using the disector. J Microsc 134(Pt 2):127–136 Rothstein JD, Ransohoff RM, Gygi SP, Antel JP, Weiner HL (2014) 24. Streit WJ (2006) Microglial senescence: does the brain's immune system Identification of a unique TGF-beta-dependent molecular and functional have an expiration date? Trends Neurosci 29(9):506–510. doi:10.1016/ signature in microglia. Nat Neurosci 17(1):131–143. doi:10.1038/nn.3599 j.tins.2006.07.001 7. Hickman SE, Kingery ND, Ohsumi TK, Borowsky ML, Wang LC, Means TK, El 25. Streit WJ, Xue QS, Tischer J, Bechmann I (2014) Microglial pathology. Acta Khoury J (2013) The microglial sensome revealed by direct RNA sequencing. Neuropathol Commun 2(1):142. doi:10.1186/s40478-014-0142-6 Nat Neurosci 16(12):1896–1905. doi:10.1038/nn.3554 26. Boche D, Perry VH, Nicoll JA (2013) Review: activation patterns of microglia and their identification in the human brain. Neuropathol Appl Neurobiol 8. Davalos D, Grutzendler J, Yang G, Kim JV, Zuo Y, Jung S, Littman DR, Dustin 39(1):3–18. doi:10.1111/nan.12011 ML, Gan WB (2005) ATP mediates rapid microglial response to local brain 27. Graeber MB (2010) Changing face of microglia. Science 330(6005):783–788. injury in vivo. Nat Neurosci 8(6):752–758. doi:10.1038/nn1472 doi:10.1126/science.1190929 9. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are 28. Mosher KI, Wyss-Coray T (2014) Microglial dysfunction in brain aging and highly dynamic surveillants of brain parenchyma in vivo. Science Alzheimer's disease. Biochem Pharmacol 88(4):594–604. doi:10.1016/ 308(5726):1314–1318. doi:10.1126/science.1110647 j.bcp.2014.01.008 10. Nelson PT, Alafuzoff I, Bigio EH, Bouras C, Braak H, Cairns NJ, Castellani RJ, Crain 29. Streit WJ, Braak H, Xue QS, Bechmann I (2009) Dystrophic (senescent) rather BJ, Davies P, Del Tredici K, Duyckaerts C, Frosch MP, Haroutunian V, Hof PR, than activated microglial cells are associated with tau pathology and likely Hulette CM, Hyman BT, Iwatsubo T, Jellinger KA, Jicha GA, Kovari E, Kukull WA, precede neurodegeneration in Alzheimer's disease. Acta Neuropathol Leverenz JB, Love S, Mackenzie IR, Mann DM, Masliah E, McKee AC, Montine TJ, 118(4):475–485. doi:10.1007/s00401-009-0556-6 Morris JC, Schneider JA, Sonnen JA, Thal DR, Trojanowski JQ, Troncoso JC, 30. Xue QS, Streit WJ (2011) Microglial pathology in down syndrome. Acta Wisniewski T, Woltjer RL, Beach TG (2012) Correlation of Alzheimer disease Neuropathol 122(4):455–466. doi:10.1007/s00401-011-0864-5 neuropathologic changes with cognitive status: a review of the literature. J 31. Lopes KO, Sparks DL, Streit WJ (2008) Microglial dystrophy in the aged and Neuropathol Exp Neurol 71(5):362–381. doi:10.1097/NEN.0b013e31825018f7 Alzheimer's disease brain is associated with ferritin immunoreactivity. Glia 11. Nelson PT,Smith CD, Abner EL,Wilfred BJ,WangWX, Neltner JH, Baker M, 56(10):1048–1060. doi:10.1002/glia.20678 Fardo DW, Kryscio RJ, Scheff SW, Jicha GA, Jellinger KA, Van Eldik LJ, 32. Streit WJ, Sammons NW, Kuhns AJ, Sparks DL (2004) Dystrophic microglia in Schmitt FA (2013) Hippocampal sclerosis of aging, a prevalent and the aging human brain. Glia 45(2):208–212. doi:10.1002/glia.10319 high-morbidity brain disease. Acta Neuropathol 126(2):161–177. 33. Smith AM, Dragunow M (2014) The human side of microglia. Trends doi:10.1007/s00401-013-1154-1 Neurosci 37(3):125–135. doi:10.1016/j.tins.2013.12.001 12. Kurushima H, Ramprasad M, Kondratenko N, Foster DM, Quehenberger O, 34. Sudduth TL, Schmitt FA, Nelson PT, Wilcock DM (2013) Neuroinflammatory Steinberg D (2000) Surface expression and rapid internalization of phenotype in early Alzheimer's disease. Neurobiol Aging 34(4):1051–1059. macrosialin (mouse CD68) on elicited mouse peritoneal macrophages. doi:10.1016/j.neurobiolaging.2012.09.012 J Leukoc Biol 67(1):104–108 35. Wirenfeldt M, Clare R, Tung S, Bottini A, Mathern GW, Vinters HV (2009) 13. da Silva RP, Gordon S (1999) Phagocytosis stimulates alternative Increased activation of Iba1(+) microglia in pediatric epilepsy patients with glycosylation of macrosialin (mouse CD68), a macrophage-specific endosomal Rasmussen's encephalitis compared with cortical dysplasia and tuberous protein. Biochem J 338(Pt 3):687–694 sclerosis complex. Neurobiol Dis 34(3):432–440. doi:10.1016/ 14. Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S (1998) j.nbd.2009.02.015 Microglia-specific localisation of a novel calcium binding protein, Iba1. 36. Taylor SE, Morganti-Kossmann C, Lifshitz J, Ziebell JM (2014) Rod microglia: Brain Res Mol Brain Res 57(1):1–9 a morphological definition. PLoS One 9(5), e97096. doi:10.1371/ 15. Schmitt FA, Nelson PT, Abner E, Scheff S, Jicha GA, Smith C, Cooper G, journal.pone.0097096 Mendiondo M, Danner DD, Van Eldik LJ, Caban-Holt A, Lovell MA, Kryscio RJ 37. Ziebell JM, Taylor SE, Cao TX, Harrison JL, Lifshitz J (2012) Rod microglia: (2012) University of Kentucky Sanders-Brown healthy brain aging volunteers: elongation, alignment, and coupling to form trains across the donor characteristics, procedures and neuropathology. Curr Alzheimer Res somatosensory cortex after experimental diffuse brain injury. J 9(6):724–733 Neuroinflammation 9. doi:10.1186/1742-2094-9-247 16. Bachstetter AD, Rowe RK, Kaneko M, Goulding D, Lifshitz J, Van Eldik LJ 38. Cao T, Thomas TC, Ziebell JM, Pauly JR, Lifshitz J (2012) Morphological and (2013) The p38alpha MAPK regulates microglial responsiveness to diffuse genetic activation of microglia after diffuse traumatic brain injury in the rat. traumatic brain injury. J Neurosci 33(14):6143–6153. doi:10.1523/ Neuroscience 225:65–75. doi:10.1016/j.neuroscience.2012.08.058 JNEUROSCI.5399-12.2013 39. Ackman JB, Siddiqi F, Walikonis RS, LoTurco JJ (2006) Fusion of microglia with 17. Bachstetter AD, Norris CM, Sompol P, Wilcock DM, Goulding D, Neltner JH, pyramidal neurons after retroviral infection. J Neurosci 26(44):11413–11422. St Clair D, Watterson DM, Van Eldik LJ (2012) Early stage drug treatment doi:10.1523/JNEUROSCI.3340-06.2006 that normalizes proinflammatory cytokine production attenuates synaptic 40. Torres-Platas SG, Comeau S, Rachalski A, Dal Bo G, Cruceanu C, Turecki G, dysfunction in a mouse model that exhibits age-dependent progression of Giros B, Mechawar N (2014) Morphometric characterization of microglial Alzheimer's disease-related pathology. J Neurosci 32(30):10201–10210. phenotypes in human cerebral cortex. J Neuroinflammation 11. doi:10.1186/ doi:10.1523/JNEUROSCI.1496-12.2012 1742-2094-11-12 18. Schmider E, Ziegler M, Danay E, Beyer L, Bühner M (2010) Is It Really 41. Beynon SB, Walker FR (2012) Microglial activation in the injured and healthy Robust? Methodology: European Journal of Research Methods for the brain: What are we really tallkine about? Practical and theoretical issues Behavioral and Social Sciences 6(4):147–151. doi:10.1027/1614-2241/a000016 associated with measurement of changes in microglia morphology. 19. Leverenz JB, Agustin CM, Tsuang D, Peskind ER, Edland SD, Nochlin D, Neuroscience 225:162–171. doi:10.1016/j.neuroscience.2012.07.029 DiGiacomo L, Bowen JD, McCormick WC, Teri L, Raskind MA, Kukull WA, 42. Nelson PT, Abner EL, Schmitt FA, Kryscio RJ, Jicha GA, Smith CD, Davis DG, Larson EB (2002) Clinical and neuropathological characteristics of Poduska JW, Patel E, Mendiondo MS, Markesbery WR (2010) Modeling the hippocampal sclerosis - A community-based study. Arch Neurol association between 43 different clinical and pathological variables and the 59(7):1099–1106. doi:10.1001/archneur.59.7.1099 severity of cognitive impairment in a large autopsy cohort of elderly 20. Nelson PT, Schmitt FA, Lin YS, Abner EL, Jicha GA, Patel E, Thomason PC, persons. Brain Pathol 20(1):66–79. doi:10.1111/j.1750-3639.2008.00244.x Neltner JH, Smith CD, Santacruz KS, Sonnen JA, Poon LW, Gearing M, Green RC, Woodard JL, Van Eldik LJ, Kryscio RJ (2011) Hippocampal sclerosis in advanced age: clinical and pathological features. Brain 134:1506–1518. doi:10.1093/brain/awr053 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Disease-related microglia heterogeneity in the hippocampus of Alzheimer’s disease, dementia with Lewy bodies, and hippocampal sclerosis of aging

Loading next page...
 
/lp/springer-journals/disease-related-microglia-heterogeneity-in-the-hippocampus-of-Y69FoS2tn2

References (46)

Publisher
Springer Journals
Copyright
Copyright © 2015 by Bachstetter et al.; licensee BioMed Central.
Subject
Biomedicine; Neurosciences; Pathology
eISSN
2051-5960
DOI
10.1186/s40478-015-0209-z
pmid
26001591
Publisher site
See Article on Publisher Site

Abstract

Introduction: Neuropathological, genetic, and biochemical studies have provided support for the hypothesis that microglia participate in Alzheimer’s disease (AD) pathogenesis. Despite the extensive characterization of AD microglia, there are still many unanswered questions, and little is known about microglial morphology in other common forms of age-related dementia: particularly, dementia with Lewy bodies (DLB) and hippocampal sclerosis of aging (HS-Aging). In addition, no prior studies have attempted to compare and contrast the microglia morphology in the hippocampus of various neurodegenerative conditions. Results: Here we studied cases with pathologically-confirmed AD (n =7), HS-Aging (n = 7), AD + HS-aging (n =4), DLB (n = 12), and normal (cognitively intact) controls (NC) (n = 9) from the University of Kentucky Alzheimer’s Disease Center autopsy cohort. We defined five microglia morphological phenotypes in the autopsy samples: ramified, hypertrophic, dystrophic, rod-shaped, and amoeboid. The Aperio ScanScope digital neuropathological tool was used along with two well-known microglial markers: IBA1 (a marker for both resting and activated microglia) and CD68 (a lysosomal marker in macrophages/microglia associated with phagocytic cells). Hippocampal staining analyses included studies of subregions within the hippocampal formation and nearby white matter. Using these tools and methods, we describe variation in microglial characteristics that show some degree of disease specificity, including, (1) increased microglia density and number in HS-aging and AD + HS-aging; (2) low microglia density in DLB; (3) increased number of dystrophic microglia in HS-aging; and (4) increased proportion of dystrophic to all microglia in DLB. Conclusions: We conclude that variations in morphologies among microglial cells, and cells of macrophage lineage, can help guide future work connecting neuroinflammatory mechanisms with specific neurodegenerative disease subtypes. Keywords: Aging, Microglia activation, Mixed dementia, Neurodegeneration, Neuroinflammation, Neuropathology * Correspondence: adam.bachstetter@uky.edu Sanders-Brown Center on Aging, University of Kentucky, 800 S. Limestone St, Lexington, KY, USA Full list of author information is available at the end of the article © 2015 Bachstetter et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 2 of 16 Introduction Materials and methods There is an increasing awareness that microglia may Human subjects have a pathogenic role in neurodegenerative diseases. Tissue samples that contained the hippocampus were ac- The discovery of genetic mutations in CD33 and quired from the UK-ADC biobank. Details of recruitment TREM2 associated with the risk of developing have been described previously [15]. Information including Alzheimer’s disease (AD) [1–4] has heightened the inter- demographic and neuropathologic data is presented est in defining microglia physiology and pathology in the (Table 1). The included cases (n = 39) represented a con- context of disease. Pio Del Rio-Hortega is credited with venience sample subdivided into groups as: NC, HS-aging, early insights into microglial pathology. He recognized AD, AD + HS-aging, or DLB. Cases represented approxi- that microglia are normally highly ramified and evenly mately age-matched sampling of the neuropathologically- distributed throughout the brain. He also noted that the defined diseases using the following criteria: AD (Braak > IV, morphology of microglia is dramatically altered in high density of neocortical amyloid plaques); isocortical response to central nervous system (CNS) pathology [5]. subtype of DLB; and HS-Aging (cell loss and gliosis out of As a molecular and functionally unique population of proportion to plaques/tangle pathology, with TDP-43 cells [6, 7], microglia exhibit a remarkable ability to pathology in the hippocampus). survey the brain and rapidly undergo a spectrum of responses to insults or tissue damage [8, 9]. The Immunostaining process by which microglia change shape, molecular Paraffin-embedded tissue sections were cut at 10-μm-thick. signature, and cellular physiology is defined as micro- Immunohistochemical (IHC) began with microwave anti- glia activation [5]. gen retrieval for 6 min (power 8) using Trilogy buffer (Cell The clinical diseaseformerlyreferredtosimplyas Marque; Rocklin, CA) for CD68 and Declere buffer (Cell “Alzheimer’sdisease” is, at the population level, a complex Marque; Rocklin, CA) for IBA1. Sections were then placed manifestation of many different brain conditions [10]. in 3% H O in methanol for 30 min. Following washes in 2 2 These age-related brain pathologies include AD (character- distilled water, sections were blocked in 5% goat serum at ized by amyloid plaques and neurofibrillary tangles), as well room temperature for 1 h. Sections were incubated in pri- as cerebrovascular disease, dementia with Lewy bodies mary antibodies IBA1 (rabbit polyclonal, 1:1,000 IHC, (DLB), and hippocampal sclerosis of aging (HS-Aging) [11]. Wako); CD68 (clone KP1) (1:50 IHC, Dako) overnight at Although each of these disorders seems to have a distinct 4°C. A biotinylated secondary antibody (Vector Laboratories) genetic, clinical, and pathological cluster of characteristics, was amplified using avidin-biotin substrate (ABC solution, to date there has not been characterization of the microglial Vector Laboratories catalog no. PK-6100), followed by color responses in these conditions. development in Nova Red (Vector Laboratories). Immuno- We sought to address questions related to microglial fluorescence (IF) staining was done following microwave morphology in neurodegenerative disease tissue: 1) Is antigen retrieval for 6 min (power 8) using Declere buffer microglia pathology seen only in the presence of amyloid (Cell Marque; Rocklin, CA) for primary antibodies to: IBA1 or tau pathology, or can it be seen in other age-related (rabbit polyclonal, 1:250 IF, Wako); and PHF-1 (1:500 IHC neurodegenerative diseases?; 2) Is there microglial re- and IF, a kind gift from Dr Peter Davies, Bronx, NY), and vi- gional heterogeneity in the hippocampus (for example, sualized using appropriate secondary antibody conjugated to gray matter only)?; and, 3) Can digital neuropathological an Alexafluor probe (1:200, Lifetechnologies) applied for 1 h. quantification detect differences in microglia activation A 0.1% solution of Sudan Black was used to reduce autofluo- in different neurodegenerative diseases? To address rescence. Slides were coverslipped using Vectashield mount- these questions, we queried well-characterized brain ing medium with DAPI (Vector Labs, Burlingame, CA). samples from the University of Kentucky Alzheimer’s Disease Center (UK-ADC) cohort. Specifically, brain tis- Quantitative image analysis sue was analyzed, incorporating multiple disease condi- Three different methods of quantitative image analysis tions, using two antibodies that react with microglia. were used in this study: 1) digital positive pixel algo- The CD68 antibody stains for a lysosomal-associated rithm, 2) digital nuclear algorithm, 3) and manual protein in macrophages/microglia and is associated with counting of IBA1 microglia (only in CA1 region). phagocytic cells [12, 13]. The IBA1 (ionized calcium Briefly, the Aperio ScanScope XT digital slidescanner binding adaptor molecule 1) antibody [14] is used widely was used to image the entire stained slide at 40x magni- as a pan marker for both resting and activated microglia. fication to create a single high-resolution digital image. Using these two widely studied microglia markers, CD68 The Aperio positive pixel count algorithm (version 9) and IBA1, we defined microglia morphologies in the was used to quantify the amount of specific staining in aged brain, including some features that show evidence the region, and the Aperio nuclear algorithm (version 9) of disease specificity. was used to determine the number of stained microglia Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 3 of 16 Table 1 Cohort demographics and numbers age at Final ApoE PMI Braak CERAD plaque diffuse neuritic diffuse Lewy case sex NFTs death MMSE alleles (h) stage stage plaques plaques bodies NC = non-demented control: mean age =86; mean MMSE = 30; Median Braak stage = 2; Median CERAD = 0 1 81 30 M 3 / 3 2.17 2 0 0 0 0 0 2 91 30 M n/a n/a 2 0 0 0 1.4 0 3 86 30 M 3 / 3 2.17 3 0 0 0 4.6 0 4 93 30 F 3 / 3 2.25 2 0 0 0 0.25 0 5 84 30 M 2 / 3 3.25 0 0 n/a n/a n/a 0 6 85 30 M 3 / 3 2 3 1 2 0.5 8 0 7 84 30 F 3 / 3 2.42 0 0 0 0 0 0 8 92 30 F 2 / 3 3.25 3 2 5 0 0.75 0 9 81 30 M 3 / 4 2 2 2 2 10.5 1 0 HS = hippocampal sclerosis of aging: mean age =87; mean MMSE = 22.7; Median Braak stage = 2; Median CERAD = 0 10 74 20 M 3 / 4 8 3 2 0 0 1.4 0 11 95 16 F 3 / 4 3.25 3 0 0 0 1 0 12 87 28 F 3 / 3 1.82 3 0 0 0 4 0 13 84 10 F n/a 2.57 2 0 0 0 0 0 14 91 29 F 2 / 3 2.87 2 1 0 0 6.8 0 15 91 30 M 3 / 3 2.83 2 0 n/a n/a n/a 0 16 88 26 M 3 / 4 2.33 0 0 0 0 0 0 AD = Alzheimer’s disease: mean age =77; mean MMSE = 11; Median Braak stage = 6; Median CERAD = 3 17 75 18 F 3 / 4 2.5 6 3 9 0.33 25 0 18 84 13 M 4 / 4 5.17 6 3 4 1.33 26.75 0 19 65 3 F 3 / 4 4.1 6 3 2.67 1.67 41 0 20 85 4 F 3 / 3 11.2 6 3 6.67 1.33 78.8 0 21 79 6 M n/a 2.08 6 3 4.33 2.33 25.6 0 22 67 11 M 2 / 3 1.75 6 3 0.33 0 2.4 + 23 82 25 M n/a 2.75 6 2 1.67 2.67 19.8 + AD + HS: mean age =91; mean MMSE = 7.8; Median Braak stage = 6; Median CERAD = 3 24 96 6 F 3 / 3 6.75 6 3 n/a n/a n/a 0 25 91 13 F 3 / 3 3 5 3 6.67 1.33 34 0 26 91 12 F 4 / 4 2.33 6 3 0 0 10.25 + 27 87 0 F 3 / 3 2.67 6 3 0 0 54.6 0 DLB = Dementia with Lewy bodies: mean age =80; mean MMSE = 17.25; Median Braak stage = 2; Median CERAD = 1 28 65 9 M 4 / 4 9.5 2 3 2 7.67 2 + 29 61 18 M 3 / 3 2 2 0 0 0 9.6 + 30 85 2 F 2 / 3 2 2 3 4 6.67 3.6 + 31 85 27 M 3 / 3 11.2 2 1 0 0 0.8 + 32 89 27 M 3 / 3 2.42 2 0 0 0 8.5 + 33 92 27 F 3 / 4 2.42 2 1 0.67 0 2 + 34 68 11 M 3 / 4 3.75 2 2 0 0 0.4 + 35 81 15 M 3 / 3 2.42 2 2 1.5 2.5 0.25 + 36 78 9 M 3 / 3 2.5 1 0 0 0 2 + 37 81 26 M 3 / 3 5.77 1 1 0 0 0 + 38 97 21 M 2 / 3 3.5 1 0 0 0 1.8 + Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 4 of 16 Table 1 Cohort demographics and numbers (Continued) 39 78 15 M 3 / 4 n/a 1 1 3.5 0.5 0 + Counts in the hippocampus of neurofibrillary tangles (NFTs), and amyloid plaques, without neurites (diffuse plaques), and with degenerating neurites (neuritic plaques) (see [42]). Abbreviations: Apo E apolipoprotein E; PMI = post-mortem interval; MMSE = mini-mental state examination; +, feature present; n/a = not available as previously described [16, 17]. The number of IBA1 Statistics microglia was counted by morphological appearance in JMP Software version 10.0 was used for statistical analysis. 5 arbitrarily placed 250 × 250 μmboxes in theCA1 re- Normality was assessed using the Shapiro-Wilk test. As gion. A researcher (coauthor ADB) blind to all samples’ there were only a few violations of normality, and ANOVA case histories conducted all data analysis. Immunofluor- is robust to such violations [18], a one-way ANOVA escence was imaged using a Nikon Eclipse 90i upright followed by a Tukey post hoc analysis was used to compare microscope equipped with a Nikon DS-Ri1 digital differences between the five groups. Mean ± SD for quanti- camera. fications are shown in Table 2. Differences between means Table 2 Summary of microglia neuropathological assessment CD68 positive pixels (Fig. 2) WM sub CA1 CA2/3 CA4 DG hipp ave NC (N = 9) 7.5 ± 4 2.5 ± 1.6 2.6 ± 2.2 3.2 ± 2.1 2.8 ± 2.2 2.4 ± 1.6 3.5 ± 2.1 HS-aging (N = 7) 8.7 ± 2.4 6 ± 3.5 5.4 ± 2.1 4.1 ± 2.1 3.4 ± 1.2 2.4 ± 1.2 5 ± 1.4 AD (N = 7) 10.6 ± 3.2 6.2 ± 2 5.3 ± 1.5 4.2 ± 2.6 3.5 ± 1.2 4.8 ± 2 5.8 ± 1 AD + HS (N = 4) 8.7 ± 3.6 5.5 ± 2.6 5.8 ± 2.1 3.2 ± 1.2 2.4 ± 1.3 2.5 ± 1.4 4.7 ± 1.2 DLB (N = 12) 7.6 ± 2.8 2.7 ± 1 2.3 ± 0.8 3.4 ± 2.3 2.7 ± 1.1 2.2 ± 0.3 3.5 ± 0.9 CD68 nuclear algorithm (Fig. 4) NC (N = 9) 54.7 ± 52.1 13.9 ± 17.9 16 ± 18.5 17 ± 18.3 20 ± 24 10.7 ± 11.3 22.1 ± 22.1 HS-aging (N = 7) 54.8 ± 45.1 34.4 ± 33.6 30.2 ± 20.9 20.4 ± 19.8 19.2 ± 11.9 8.1 ± 7.8 27.8 ± 21.2 AD (N = 7) 100.2 ± 68.4 45.8 ± 35.8 29.2 ± 16.8 16.4 ± 7.5 17.3 ± 13.3 26.3 ± 18.1 39.2 ± 22.4 AD + HS (N = 4) 32.4 ± 15.8 27.2 ± 22.4 28.7 ± 26.6 15.4 ± 14.4 8 ± 6.8 3.8 ± 5.2 19.2 ± 13.7 DLB (N = 12) 53.3 ± 35 11.8 ± 9 11.6 ± 7.7 15.6 ± 11.1 13.3 ± 8.1 8.9 ± 5.9 19.1 ± 10.1 IBA1 nuclear algorithm (Fig. 5) NC (N = 9) 101.9 ± 30.2 66.5 ± 14.3 60.8 ± 22.9 74 ± 11.7 82.7 ± 16.8 80.5 ± 18.7 77.7 ± 11.7 HS-aging (N = 7) 143.7 ± 80.8 84.7 ± 39.4 122.1 ± 63.5 142.8 ± 37.3 110.9 ± 50.3 92.3 ± 39.1 116.1 ± 45.4 AD (N = 7) 87.8 ± 55.4 84.1 ± 44.2 64.2 ± 16.2 75.5 ± 19.2 43.2 ± 27.2 50.5 ± 37.2 67.5 ± 26.7 AD + HS (N = 4) 97.9 ± 25.8 116.4 ± 74.9 108.2 ± 42.1 132.4 ± 64.6 94.3 ± 47.1 98.9 ± 47.1 108 ± 41.9 DLB (N = 12) 103.2 ± 55.6 69.3 ± 35.6 70 ± 32 74.2 ± 46.7 83.8 ± 48 74.5 ± 43.9 79.2 ± 40.7 IBA1 positive pixels (Fig. 6) NC (N = 9) 2.8 ± 0.8 2.6 ± 0.9 2.4 ± 0.7 3.1 ± 0.8 3.3 ± 1 3.1 ± 0.8 2.9 ± 0.7 HS-aging (N = 7) 4 ± 1.8 2.5 ± 0.9 3.2 ± 1.1 3.9 ± 0.8 3.5 ± 1.3 3.1 ± 1.1 3.3 ± 1 AD (N = 7) 2.7 ± 1.2 2.5 ± 1.1 2.2 ± 0.6 3.1 ± 1 2 ± 1.1 2.1 ± 0.8 2.4 ± 0.7 AD + HS (N = 4) 3.6 ± 0.8 3.4 ± 2.1 3.6 ± 1.1 3.9 ± 1.2 3.1 ± 0.7 3.3 ± 1 3.5 ± 1 DLB (N = 12) 2.8 ± 0.8 1.9 ± 0.6 1.9 ± 0.5 2.3 ± 0.8 2.6 ± 0.9 2.5 ± 0.9 2.3 ± 0.7 Morphological assessment of IBA1 microglia in CA1 region (Fig. 7) ramified hypertrophic dystrophic rod-shaped amoeboid total NC (N = 9) 16.9 ± 9.8 3.7 ± 7.1 4.4 ± 3.9 2.1 ± 4.1 3.7 ± 5 30.8 ± 10.3 HS-aging (N = 7) 2.6 ± 2.6 13.8 ± 11.9 32.2 ± 22 2.4 ± 5.1 8.8 ± 8.8 59.9 ± 29.8 AD (N = 7) 19.7 ± 10.4 4.5 ± 6.1 11.2 ± 9.7 2.6 ± 2.7 4.5 ± 3.6 42.4 ± 11.5 AD + HS (N = 4) 9.8 ± 8.3 20.2 ± 13.7 19.8 ± 11.2 2.6 ± 3.2 14.6 ± 12.4 67 ± 15.8 DLB (N = 12) 9.3 ± 6.5 2.5 ± 3.6 13.9 ± 9.3 3.9 ± 4 1.9 ± 2.6 31.3 ± 8.2 Values represent mean ± SD for the quantification of CD68 and IBA1 immunohistochemistry. Data is plotted in the indicated figures Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 5 of 16 were considered significant at p < 0.05. Heatmaps were gen- of the cerebral neocortex. In pure DLB, there are low levels erated using JMP Software version 10.0. All other graphs of amyloid-β pathology or NFTs, as shown in Table 1. were generated using GraphPad Prism software version 6.0, Primary goals of this study were to assess regional with values expressed as mean ± SEM. microglia heterogeneity and to exploit the ability of digital neuropathological quantification to detect in differences microglial morphometry when cases are stratified accord- Results ing to their neurodegenerative diseases. Six regions of Five groups of cases (Table 1) were pathologically-confirmed interest (ROI) were identified by dividing the hippocampal as either AD (n=7), HS-Aging (n =7), AD +HS-aging (n = formation into the dentate gyrus (DG), the cornu ammo- 4), DLB (n = 12), and NC (n = 9). HS-aging and DLB cases nis (CA) areas (CA1, CA2/3, and CA4), the subiculum were included in this study to determine if there is disease (sub), and the adjacent white matter (WM) (Fig. 1). Repre- specificity in microglia pathology and to provide the first sentative examples of the ROIs are shown in Fig. 1. quantitative analysis of microglia in HS-Aging. Pure HS-aging cases lacked substantial additional patholo- Pattern of CD68 staining in the hippocampus of autopsy gies AD-type pathology, or Lewy bodies [19–21], as cases shown in Table 1. The neuropathological changes asso- Quantification of the CD68 positive pixels is shown in ciated with neocortical/diffuse Lewy body disease in- Fig. 2. By a one-way ANOVA a significant effect of dis- clude, by definition, α-synuclein immunoreactive neuronal ease status was found sub (Fig. 2c; F = 6.3001; p = 4,38 inclusions (Lewy bodies) and processes in multiple portions 0.0007), CA1 (Fig. 2d; F = 8.0944; p < 0.001), DG 4,38 Fig. 1 Regions of interest used for microglia analysis. A representative hippocampus is shown for the five neuropathological diagnoses included in this study. The outlines illustrate the boundaries used in identifying the following brain regions: white matter (WM), subiculum (sub), the cornu ammonis(CA)areas,CA1,CA2/3,CA4, and thedentate gyrus(DG). TheROIsshowninthe figure arenot theactualROIsusedfor analysis,assomeofthe ROIs (WM and sub) could not be included in the image frame, as the brain region was larger than the image frame Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 6 of 16 Fig. 2 Digital neuropathological quantification using positive pixel algorithm of CD68 immunostaining in the hippocampus of autopsy cases. Representative example of (a) CD68 staining and a digitally generated mark-up showing the ability of the positive pixel algorithm to detect the staining. Digital neuropathological quantification of the CD68 staining using the positive pixel algorithm is shown for the (b) WM, (c) sub, (d) CA1, (e) CA2/3, (f) CA4, (g) DG, and for the (h) average of the hippocampal formation. Circles represent an individual case, with mean and SEM shown § ‡ for the group. Statistical comparisons: *p <0.05 compared to AD cases. p<0.05comparedtoHS-agingcases. p < 0.05 compared to AD + HS-aging cases. (i) Heatmap summarizes the results shown in (b-h) (also see Table 2) (Fig. 2g; F = 5.3332; p = 0.0019), and in the average regional and disease-specific heterogeneity in the stain- 4,38 of the six regions in the hippocampus formation ing (Fig. 3). Of note is a large round cell type that can (Fig. 2h; F = 4.3221; p = 0.0062). No significant effect be found in areas of high density staining as shown in 4,38 was found by a one-way ANOVA in WM (Fig. 2b), Fig. 3b-c. Interestingly, just distal to the very intense + + CA2/3, (Fig.2e),orCA4 (Fig.2f).HS-aging, AD,and accumulation of CD68 cells, the CD68 staining was AD + HS-aging were found to have significantly more unremarkable, with a few ramified microglia (Fig. 3d). + + CD68 staining in the CA1 region compared to NC or Quantification of the number of large round CD68 DLB cases (Fig.2d).However,there was nosignificant cells was done using the nuclear algorithm, by adjusting differenceamong thethree disease conditions(HS- the algorithm to detect only the large round cells as aging, AD, and AD + HS-aging) in the CA1 region shown inFig.4.Incomparisontodesignbased stereo- (Fig. 2d). Interestingly, we found significantly more logical methods, limitations of the nuclear algorithm CD68 staining in the DG of AD cases compared to the include an inability to provide an estimate of the total other four groups (Fig. 2g). When averaged across the number of microglia, because of a lack of 3- six-hippocampal formation sub regions, the AD cases dimensional volume measurements [22, 23]. Limita- were found to have significantly more CD68 staining tions notwithstanding, results of the nuclear algorithm compared to NC or DLB groups. Overall, the greatest were similar to the positive pixel algorithm, with the CD68 staining was seen in the WM, as is evident by HS-aging and AD groups having the greatest number the heatmap summary of the CD68 positive pixel of CD68 cells (Table 2). As shown by the heatmap, the + + analysis (Fig. 2i). A survey of the CD68 staining in the greatest number of CD68 cells was found in the WM six-hippocampal formation regions illustrates the of AD cases (Fig. 4). Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 7 of 16 + + Fig. 3 Survey of CD68 staining in the hippocampus of autopsy cases. (a) Representative examples of CD68 staining pattern in the brain regions analyzed by digital neuropathological analysis. (b) Low power photomicrograph of hippocampus of a DLB individual (case #36) highlights an area of intense staining (blue arrow) shown in (c), and an area of low CD68 staining (black arrow) in a nearby region (d) Digital quantification of IBA1 staining in the of disease status was found in the CA1 region (Fig. 5d; hippocampus of autopsy cases F = 3.9914; p = 0.0092), CA2/3 region (Fig. 5e; F = 4,38 4,38 Quantification of the number of IBA1 cells by the nuclear 5.8525; p = 0.0011), and in the CA4 (Figure 5f F = 4,38 algorithm is shown in Fig. 5. A representative example of 2.6929; p = 0.0473). No significant effect was found by a the ability of the algorithm to detect individual cells is one-way ANOVA in WM (Figure 5b), sub (Fig. 5c), DG shown in Fig. 5a. By a one-way ANOVA, a significant effect (Fig. 5g), or in the average of the six regions in the Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 8 of 16 Fig. 4 Digital neuropathological quantification using nuclear algorithm of number of large round CD68 cells in the hippocampus of autopsy cases. Representative example of (a) CD68 staining and a digitally generated mark-up showing the ability of the nuclear algorithm to detect the staining of the large round cells, but not smaller cells or processes. Digital neuropathological quantification of the CD68 staining using the nuclear algorithm to detect the staining is shown for the (b)WM, (c)sub (F = 2.9934; p = 0.0321), (d)CA1,(e)CA2/3,(f)CA4,(g)DG (F = 4.3393; p =0.0061), 4,38 4,38 and for the (h) average of the hippocampal formation. Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: *p <0.05 compared to AD cases. (i) Heatmap summarizes the results shown in (b-h) (also see Table 2) hippocampus formation (Fig. 5h). In the CA1 region, the microglia density and cell number were measured irrespect- HS-aging had an increased number of IBA1 cells com- ive of the microglia morphology. For example, a striking pared to NC, AD or DLB. As shown by the heatmap sum- pattern of IBA1 microglia morphology is the rod-shaped mary, a similar pattern of increased number of IBA1 microglia, which were readily apparent in a subset of cases. microglia was found in the HS-aging and AD + HS-aging As shown in Fig. 7b-c, rod-shaped microglia are character- groups compared to the NC, AD, or DLB groups ized by a narrow cell body with a few planar processes. The (Fig. 5i). Quantification of the IBA1 positive pixels (Table 2) rod-shaped microglia could be found as individual cells also showed a similar pattern of increased IBA1 staining (Fig. 7b), or as long and thin groups of cells that may have in the HS-aging and AD + HS-aging groups compared to fused(Fig.7b and c andFig.8).Theappearanceofmicro- the NC, AD, or DLB groups (Fig. 6). glia with polarized and parallel processes suggested that the microglia could be following neurites—possibly, degenerat- IBA1 microglia morphology in the hippocampus of ing axons or neurons themselves. To test the possibility that autopsy cases microglia could be surrounding degenerating neuronal pro- An examination (Fig. 7a) of the IBA1 microglia in the six cesses, double label immunofluorescence was performed ROIs in the five neuropathologic groups showed remark- for microglia (IBA1) and NFTs (PHF1). Fig. 8a shows abun- + + able heterogeneity in microglia density, as captured by the dant PHF1 staining and IBA1 rod-shaped microglia in digital neuropathological quantification. There was also het- theCA1 region of an AD individual (case#23).Wefound + + erogeneity in IBA1 microglia morphology, which was un- no evidence of systematic overlap of PHF1 neurites and derappreciated in the digital neuropathological analysis, as IBA1 rod-shaped microglia, as shown in Fig. 8b. Rather, Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 9 of 16 Fig. 5 Digital neuropathological quantification using nuclear algorithm of number of IBA1 cells in the hippocampus of autopsy cases. Representative example of (a) IBA1 staining and a digitally generated mark-up showing the ability of the nuclear algorithm to detect six stained cells. Digital neuropathological quantification of the IBA1 staining using the nuclear algorithm is shown for the (b) WM, (c) sub, (d) CA1, (e) CA2/ 3, (f) CA4, (g) DG, and for the (h) average of the hippocampal formation.. Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: p < 0.05 compared to HS-aging cases. (i) Heatmap summarizes the results shown in (b-g) (also see Table 2) long trains of rod-shaped microglia could sometimes be (Fig. 10a), to allow measurement of changes in the micro- seen to run parallel to and between PHF1 neurons but did glia classes associated with the five neurodegenerative dis- not co-localize with the PHF1 staining (Fig. 8c). In this ex- ease groups. The five classes of microglia morphologies ample, the tip of the rod-shaped microglia was near (but included: 1) ramified microglia, which have a ‘surveying’ not within) a PHF1 structure, and the IBA-1 immunoreac- non-reactive microglia morphological appearance, with thin tive structure appeared to be a fusion / cluster of multiple highly branched processes [5, 26]; 2) hypertrophic microglia cells with 5 clearly visible DAPI nuclei (Fig. 8d). (often called activated microglia), which have become en- Another pattern of microglia morphology observed larged, hyper-ramified or may have short thick processes [5, was the dystrophic / degenerating microglia, which over- 26]; 3) dystrophic microglia, with processes that are spher- lapped morphologically with cells that have been de- oidal, beaded, de-ramified, or fragmented [24–26]; 4) rod- scribed to have processes that are spheroidal, beaded, shaped microglia, characterized by a narrow cell body with de-ramified, or fragmented [24, 25]. Examples of dys- a few planar processes [27, 26]; and 5) amoeboid microglia, trophic / degenerating microglia are shown in Fig. 9. In with an enlarged cell body with few to no processes [5, 26]. AD (Fig. 9a) and DLB (Fig. 9b), for example, the dys- CD68 staining could clearly identify cells with an amoeboid trophic / degenerating microglia had very thin processes morphology, and to a lesser extent cells with a ramified that are beaded and fragmented. In HS-aging (Fig. 9c) morphology. In contrast, IBA1 staining was useful to iden- and AD + HS-aging (Fig. 9d), dystrophic microglia tify all five microglia morphologies. Therefore, IBA1 stain- morphology was more striking, and the processes of the ing was used to quantify the distribution in the microglia microglia were beaded and tortuous. morphology according to these five subtypes of microglial The remarkable diversity in the microglia morphology shapes. Focusing on the CA1 region of the hippocampus, led us to carefully review and categorize the morphological the numberofeachofthe five morphological classesof appearances of the microglia into five distinct classes IBA1 microglia was counted in five randomly placed Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 10 of 16 Fig. 6 Digital neuropathological quantification using positive pixel algorithm of IBA1 immunostaining in the hippocampus of autopsy cases. Representative example of (a) IBA1 staining and a digitally generated mark-up showing the ability of the positive pixel algorithm to detect the staining. Digital neuropathological quantification of the IBA1 staining using the positive pixel algorithm is shown for the (b) WM, (c) sub, (d) CA1 (F =5.0943; p =0.0025), (e)CA2/3 (F =4.8888; p = 0.0032), (f)CA4, (g) DG,, and for the (h) average of the hippocampal formation (F =3.0201; 4,38 4,38 4,38 p = 0.0311). Circles represent an individual case, with mean and SEM shown for the group. Statistical comparisons: *p <0.05 compared to § ‡ AD cases. p < 0.05 compared to HS-aging cases. p < 0.05 compared to AD + HS-aging cases. (i) Heatmap summarizes the results in (b-h) (also see Table 2) and evenly distributed 250x250μm regions of interest (p = 0.0035) cases. HS-aging cases had more total (ROI). HS-aging cases had fewer ramified microglia microglia than NC (p = 0.0072) or DLB (p = 0.0048) than NC (p = 0.0091) or AD (p = 0.0027) cases (Fig. 10b, cases (Fig. 10g, Table 2). As the total number of micro- Table 2). HS-aging and AD + HS-aging had the most hyper- glia was found to be altered in the different groups, trophic microglia. AD + HS-aging cases had more hyper- each of the five microglia classifications was plotted as trophic microglia than NC (p = 0.0132), AD (p = 0.0270), or a percentage of the total number of microglia (Fig. 10h) DLB (p = 0.0044) cases, and HS-aging cases had more to help visualize the microglia morphology distribu- hypertrophic microglia than DLB (p = 0.0140) cases tions within and among the different diseases. (Fig. 10c, Table 2). HS-aging cases had more dystrophic microglia than NC (p = 0.0005), AD (p = 0.0193), or Discussion DLB (p = 0.0225) cases (Fig. 10d, Table 2). Quantification of The present study underscores the rich diversity of micro- rod-shaped microglia identified a subset of cases with glial morphologies in the hippocampus of the human brain abundant rod-shaped microglia; however, the cases that maychangeaccording to thediseasesofaging.Weob- were not specific to a disease group (Fig. 10e, Table 2). served regional heterogeneity in the hippocampal formation + + AD + HS-aging cases had more amoeboid microglia in the density and number of IBA1 and CD68 microglia. than NC (p = 0.0428), or DLB (p = 0.0085) cases We also observed five morphologically-defined classes of (Fig. 10f, Table 2). The total number of microglia in the IBA1 labeled microglia: ramified, hypertrophic, dystrophic, CA1 region, regardless of morphology, was greatest in rod-shaped, and amoeboid (Fig. 10). Our observations pro- HS-aging and AD + HS-aging. AD + HS-aging cases vide evidence for subclasses of microglial morphologies that had more total microglia than NC (p = 0.046), or DLB are seen in particular neurodegenerative diseases. The data Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 11 of 16 + + Fig. 7 Survey of IBA1 staining in the hippocampus of autopsy cases. (a) Representative examples of IBA1 staining pattern in the brain regions analyzed by digital neuropathological analysis(b) A low powered photomicrograph shows the widespread distribution of rod shaped microglia in the CA1 region of a DLB individual (case #34). Long trains of microglia (highlighted by blue arrows) are shown at higher magnification in (c). provide at least some support for disease-specific microglia neuropathological quantification to measure changes in pathology in age-related dementias. human microglia activation and compare directly the A primary goal of our project was to determine if microglia response in the different neurodegenerative digital neuropathological quantification could detect diseases, and the first to assess microglia in HS-aging disease-specific changes in IBA1 and CD68 labeled cases. The digital neuropathological quantification was microglia activation. This is the first study to use digital able to detect regional differences in IBA1 and CD68 Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 12 of 16 + + Fig. 8 Lack of localization of IBA rod-shaped microglia to PHF1 neurons in an AD individual (case #23). (a) A low powered photomicrograph shows the distribution of rod-shaped microglia next to PHF1 cells. (b) A linear group of rod-shaped microglia is shown at a higher magnification. (c) A second example of rod-microglia, where the microglia run parallel and between PHF1+ neurons. (d)Ofnote, thepolar endof the rod-microglia(white arrow)was foundtohave5DAPI nuclei staining associated with the neuropathological diagno- cases, suggesting that the HS-aging pattern of microglia sis. Specifically, we found increased IBA1 and CD68 staining is dominant over the AD pattern, and that staining in the HS-aging and AD + HS-aging cases. there is not a robust additive effect of the two patholo- Interestingly, the spatial pattern and magnitude of the gies. Thus, results of the digital neuropathological changes in IBA1 and CD68 staining were remarkably quantification clearly show a pattern of microglia acti- similar between the HS-aging and AD + HS-aging vation associated with a specific neurodegenerative Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 13 of 16 + + Fig. 9 Dystrophic IBA1 microglia in the hippocampus. Examples of IBA1 dystrophic microglia in the CA1 region of AD individual (a; case #20), DLB individual (b; case #33), HS-aging individual (c; case #15), and AD + HS-aging individual (d; case #27). Scale bar is 25 μm disease, but overall the quantification provided only of function; see [25, 28]). Support for the hypofunctional modest sensitivity, with limited diagnostic potential, to (as opposed to activated) microglia model is largely based separate AD from HS-aging. To determine the repro- on morphological examination of IBA1-stained microglia in ducibility of the digital quantification, 12 of 39 cases se- autopsy samples from aged humans [29–32], as currently lected at random were replicated in an independent there are no specific markers that recognize only degenerat- experiment. Even with the modest sample size, a com- ing/dystrophic microglia. In addition, the dystrophic micro- parison of the number of IBA1 positive pixels in the glia phenotype seen in humans is largely absent in rodent CA1 regions between the two independent experiments models [25]. This may reflect intrinsic differences in human resulted in a R = 0.7. The results of the replication microglia [33], or may reflect limitations in the current ani- study support the use of digital neuropathological quan- mal models. We found that aged individuals without de- tification as a relatively accurate, unbiased, quantitative and mentia were more likely to have ramified microglia than efficient means of neuropathological assessment. In the fu- individuals with dementia (AD, HS-aging, AD + HS-aging, ture, development of algorithms that can detect the differ- or DLB). Moreover, the present study confirmed that dys- ent microglia phenotypes (Fig.10) should greatlyimprove trophic microglia are found in aged individuals and in in- the potential of this approach to detect disease-related creased numbers in aged individuals with three distinct changes in microglia morphology, until specific molecular forms of dementia (AD, HS-aging, and DLB). Our results markers that recognize the different microglia morpho- provide an independent confirmation of the presence of logical states are available. dystrophic microglia described by the Streit laboratory The long-standing view that microglia become activated [29–32]. Research at our center has previously shown and promote neuroinflammation in neurodegenerative dis- differences in the M1/M2 microglia phenotype between ease (toxic gain of function) has been challenged recently mild AD and end-stage AD [34], supporting changes in by the concept of the dystrophic / diseased microglia (loss the temporal dynamic of the microglia response to Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 14 of 16 Fig. 10 Disease specific patterns in IBA1 microglia morphology. (a) Representation of microglia morphologies seen in the hippocampus of aged individuals. The number of microglia was quantified at 40x magnification in five 250 x 250 μm regions of interest (ROIs) that were randomly placed and evenly spaced in the CA1 region. Following the classification shown in (a), IBA microglia were classified as either (b) ramified (F = 5.3533; p = 0.0019), 4,38 (c)hypertrophic(F = 5.5082; p = 0.0016), (d) dystrophic (F = 5.7249; p = 0.0012), (e) rod-shaped, or (f) amoeboid (F = 3.9836; p = 0.0093). (g)The 4,38 4,38 4,38 number of microglia (F = 7.2694; p = 0.0002) in the five classifications was summed to get the total number of microglia. The gray circles in 4,38 (b-g) represent the average number of microglia per mm for an individual case, with mean and SEM shown for each group (see also Table 2). § ‡ Statistical comparisons: p < 0.05 comparedtoHS-agingcases. p < 0.05 compared to AD + HS-aging cases. (h)Asthe totalnumberof microglia significantly varied by group, the number of microglia in each of the five classifications was plotted as a percent of the total number of microglia to illustrate the disease-related patterns in microglia morphology (also see Table 2) Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 15 of 16 varying degrees of neuropathology. Still, the temporal able to provide average cell body size and roundness, along dynamic of microglia in humans over the course of the with the number of processes, process length and volume lifespan has not been defined and is not fully testable occupied by the processes [40].That study didnot include through autopsy (cross-sectional) studies. This is a vital any samples without neurologic disease, and therefore area for future investigation. underestimated the heterogeneity in microglia morphology; The current study highlights the importance of for example, they did not describe rod-shaped microglia. morphology-based readout of cell activity. Rod-shaped Using a similar approach, others have attempted to define microglia are a particularly fascinating microglia phenotype, classes of microglia morphology, such as the rod-shaped which was first described by Nissl more than 100 years ago microglia, by calculating celllengthto cellwidth andthe (reviewed in: [27]). Rod-shaped microglia have been de- number of polar vs. planar branches [36]. Moreover, others scribed clinically in neurosyphilis, subacute sclerosing have proposed digital 3D reconstruction of the microglia as panencephalitis, lead encephalopathy, viral encephalitis a means to quantify the microglia morphology [41]. How- including HIV-1, and Rasmussen's encephalitis [27, 35]; ever, before microglia morphological assessment can be- however, there are few modern reports of rod-shaped come standard practice in characterizing the microglia microglia in the clinical literature. In experimental pathology, a consensus must be established on what defines models, rod-shaped microglia have been best described different microglia morphologies, as there is currently no following traumatic brain injury [36–38, 16], where a consensus-based agreement on definitions, or terminology diffuse braininjurywill causethe rapid (by 6 h) for the specific classes of microglia morphology. Our study polarization of microglia to follow along neuronal pro- provides a first step towards this goal and will hopefully cesses. It has been shown previously in rats that micro- provide a framework to move the field forward in this glia will fuse specifically to the apical dendrite of direction. neurons infected with a retrovirus, but not to un- Conflict of interests infected neurons [39]. It is not clear if fusion is occur- The authors declare that they have no competing interests ring in the case of rod-shaped microglia in the current Acknowledgements study. Beyond these few reported observations, little is We are profoundly grateful to all of the study participants who make this known mechanistically about the chemoattractant sig- research possible. The corresponding author, Adam Bachstetter, PhD, had full nals that drive formation of rod-shaped microglia, or access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Research reported in this about the specific functions of the rod-shaped micro- publication was supported by National Institutes of Health under award glia in relation to the neuron. We found that rod- numbers P30 AG028383, K99 AG044445. The content is solely the shaped microglia could be proximal and parallel to responsibility of the authors and does not represent the official views of the National Institutes of Health. PHF1 neurons/axons, but the rod-shaped microglia did not appear to fuse with or engulf the PHF1 struc- Author details tures. Rod-shaped microglia were present in approxi- Sanders-Brown Center on Aging, University of Kentucky, 800 S. Limestone St, Lexington, KY, USA. Department of Anatomy and Neurobiology, mately 60% of cases included in this study, but were University of Kentucky, Lexington, KY, USA. Department of Neurology, most abundant in a subset of cases. Review of the case University of Kentucky, Lexington, KY, USA. Department of Pathology and histories of individuals with abundant rod-shaped Laboratory Medicine, Division of Neuropathology, University of Kentucky, Lexington, KY, USA. Department of Epidemiology, University of Kentucky, microglia did not identify any obvious commonalities. Lexington, KY, USA. Department of Biostatistics, University of Kentucky, A goal for future studies will be to identify a larger Lexington, KY, USA. Department of Statistics, University of Kentucky, sample of cases with abundant rod-shaped microglia to Lexington, KY, USA. distinguish clinical-pathological correlations, as a first Received: 1 May 2015 Accepted: 4 May 2015 step in defining mechanistically the functions of this mysterious cell type. References A limitation of this study is the subjective criteria that 1. Malik M, Simpson JF, Parikh I, Wilfred BR, Fardo DW, Nelson PT, Estus S (2013) were used to classify microglia into one of the five morpho- CD33 Alzheimer's risk-altering polymorphism, CD33 expression, and exon 2 logical categories (Fig. 10). This approach requires an ex- splicing. J Neurosci 33(33):13320–13325. doi:10.1523/JNEUROSCI.1224-13.2013 2. Griciuc A, Serrano-Pozo A, Parrado AR, Lesinski AN, Asselin CN, Mullin K, perimenter capable of discerning differences in microglia Hooli B, Choi SH, Hyman BT, Tanzi RE (2013) Alzheimer's disease risk gene morphology. The approach also imparts bias and potential CD33 inhibits microglial uptake of amyloid beta. Neuron 78(4):631–643. for experimenter error. The current study demonstrates doi:10.1016/j.neuron.2013.04.014 3. Jonsson T, Stefansson H, Steinberg S, Jonsdottir I, Jonsson PV, Snaedal J, that there is a great diversity in the microglia morphology Bjornsson S, Huttenlocher J, Levey AI, Lah JJ, Rujescu D, Hampel H, Giegling in humans, which is underappreciated, as this diversity is I, Andreassen OA, Engedal K, Ulstein I, Djurovic S, Ibrahim-Verbaas C, Hofman A, largely absent from animal models. There have been prior Ikram MA, van Duijn CM, Thorsteinsdottir U, Kong A, Stefansson K (2013) Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med attempts to operationalize morphological changes in micro- 368(2):107–116. doi:10.1056/NEJMoa1211103 glia. A recent study reconstructed microglia from mice and 4. Guerreiro R, Wojtas A, Bras J, Carrasquillo M, Rogaeva E, Majounie E, humans using computer-based tracing systems, and was Cruchaga C, Sassi C, Kauwe JS, Younkin S, Hazrati L, Collinge J, Pocock J, Bachstetter et al. Acta Neuropathologica Communications (2015) 3:32 Page 16 of 16 Lashley T, Williams J, Lambert JC, Amouyel P, Goate A, Rademakers R, 21. Dickson DW, Davies P, Bevona C, Vanhoeven KH, Factor SM, Grober E, Morgan K, Powell J, St George-Hyslop P, Singleton A, Hardy J, Alzheimer Aronson MK, Crystal HA (1994) Hippocampal sclerosis - a common Genetic Analysis G (2013) TREM2 variants in Alzheimer's disease. N Engl pathological feature of dementia in very old (greater-than-or-equal-to-80 years of J Med 368(2):117–127. doi:10.1056/NEJMoa1211851 age) humans. Acta Neuropathol 88(3):212–221 5. Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (2011) Physiology of 22. West MJ (1993) New stereological methods for counting neurons. Neurobiol microglia. Physiol Rev 91(2):461–553. doi:10.1152/physrev.00011.2010 Aging 14(4):275–285 6. Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, 23. Sterio DC (1984) The unbiased estimation of number and sizes of arbitrary Koeglsperger T, Dake B, Wu PM, Doykan CE, Fanek Z, Liu L, Chen Z, particles using the disector. J Microsc 134(Pt 2):127–136 Rothstein JD, Ransohoff RM, Gygi SP, Antel JP, Weiner HL (2014) 24. Streit WJ (2006) Microglial senescence: does the brain's immune system Identification of a unique TGF-beta-dependent molecular and functional have an expiration date? Trends Neurosci 29(9):506–510. doi:10.1016/ signature in microglia. Nat Neurosci 17(1):131–143. doi:10.1038/nn.3599 j.tins.2006.07.001 7. Hickman SE, Kingery ND, Ohsumi TK, Borowsky ML, Wang LC, Means TK, El 25. Streit WJ, Xue QS, Tischer J, Bechmann I (2014) Microglial pathology. Acta Khoury J (2013) The microglial sensome revealed by direct RNA sequencing. Neuropathol Commun 2(1):142. doi:10.1186/s40478-014-0142-6 Nat Neurosci 16(12):1896–1905. doi:10.1038/nn.3554 26. Boche D, Perry VH, Nicoll JA (2013) Review: activation patterns of microglia and their identification in the human brain. Neuropathol Appl Neurobiol 8. Davalos D, Grutzendler J, Yang G, Kim JV, Zuo Y, Jung S, Littman DR, Dustin 39(1):3–18. doi:10.1111/nan.12011 ML, Gan WB (2005) ATP mediates rapid microglial response to local brain 27. Graeber MB (2010) Changing face of microglia. Science 330(6005):783–788. injury in vivo. Nat Neurosci 8(6):752–758. doi:10.1038/nn1472 doi:10.1126/science.1190929 9. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are 28. Mosher KI, Wyss-Coray T (2014) Microglial dysfunction in brain aging and highly dynamic surveillants of brain parenchyma in vivo. Science Alzheimer's disease. Biochem Pharmacol 88(4):594–604. doi:10.1016/ 308(5726):1314–1318. doi:10.1126/science.1110647 j.bcp.2014.01.008 10. Nelson PT, Alafuzoff I, Bigio EH, Bouras C, Braak H, Cairns NJ, Castellani RJ, Crain 29. Streit WJ, Braak H, Xue QS, Bechmann I (2009) Dystrophic (senescent) rather BJ, Davies P, Del Tredici K, Duyckaerts C, Frosch MP, Haroutunian V, Hof PR, than activated microglial cells are associated with tau pathology and likely Hulette CM, Hyman BT, Iwatsubo T, Jellinger KA, Jicha GA, Kovari E, Kukull WA, precede neurodegeneration in Alzheimer's disease. Acta Neuropathol Leverenz JB, Love S, Mackenzie IR, Mann DM, Masliah E, McKee AC, Montine TJ, 118(4):475–485. doi:10.1007/s00401-009-0556-6 Morris JC, Schneider JA, Sonnen JA, Thal DR, Trojanowski JQ, Troncoso JC, 30. Xue QS, Streit WJ (2011) Microglial pathology in down syndrome. Acta Wisniewski T, Woltjer RL, Beach TG (2012) Correlation of Alzheimer disease Neuropathol 122(4):455–466. doi:10.1007/s00401-011-0864-5 neuropathologic changes with cognitive status: a review of the literature. J 31. Lopes KO, Sparks DL, Streit WJ (2008) Microglial dystrophy in the aged and Neuropathol Exp Neurol 71(5):362–381. doi:10.1097/NEN.0b013e31825018f7 Alzheimer's disease brain is associated with ferritin immunoreactivity. Glia 11. Nelson PT,Smith CD, Abner EL,Wilfred BJ,WangWX, Neltner JH, Baker M, 56(10):1048–1060. doi:10.1002/glia.20678 Fardo DW, Kryscio RJ, Scheff SW, Jicha GA, Jellinger KA, Van Eldik LJ, 32. Streit WJ, Sammons NW, Kuhns AJ, Sparks DL (2004) Dystrophic microglia in Schmitt FA (2013) Hippocampal sclerosis of aging, a prevalent and the aging human brain. Glia 45(2):208–212. doi:10.1002/glia.10319 high-morbidity brain disease. Acta Neuropathol 126(2):161–177. 33. Smith AM, Dragunow M (2014) The human side of microglia. Trends doi:10.1007/s00401-013-1154-1 Neurosci 37(3):125–135. doi:10.1016/j.tins.2013.12.001 12. Kurushima H, Ramprasad M, Kondratenko N, Foster DM, Quehenberger O, 34. Sudduth TL, Schmitt FA, Nelson PT, Wilcock DM (2013) Neuroinflammatory Steinberg D (2000) Surface expression and rapid internalization of phenotype in early Alzheimer's disease. Neurobiol Aging 34(4):1051–1059. macrosialin (mouse CD68) on elicited mouse peritoneal macrophages. doi:10.1016/j.neurobiolaging.2012.09.012 J Leukoc Biol 67(1):104–108 35. Wirenfeldt M, Clare R, Tung S, Bottini A, Mathern GW, Vinters HV (2009) 13. da Silva RP, Gordon S (1999) Phagocytosis stimulates alternative Increased activation of Iba1(+) microglia in pediatric epilepsy patients with glycosylation of macrosialin (mouse CD68), a macrophage-specific endosomal Rasmussen's encephalitis compared with cortical dysplasia and tuberous protein. Biochem J 338(Pt 3):687–694 sclerosis complex. Neurobiol Dis 34(3):432–440. doi:10.1016/ 14. Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S (1998) j.nbd.2009.02.015 Microglia-specific localisation of a novel calcium binding protein, Iba1. 36. Taylor SE, Morganti-Kossmann C, Lifshitz J, Ziebell JM (2014) Rod microglia: Brain Res Mol Brain Res 57(1):1–9 a morphological definition. PLoS One 9(5), e97096. doi:10.1371/ 15. Schmitt FA, Nelson PT, Abner E, Scheff S, Jicha GA, Smith C, Cooper G, journal.pone.0097096 Mendiondo M, Danner DD, Van Eldik LJ, Caban-Holt A, Lovell MA, Kryscio RJ 37. Ziebell JM, Taylor SE, Cao TX, Harrison JL, Lifshitz J (2012) Rod microglia: (2012) University of Kentucky Sanders-Brown healthy brain aging volunteers: elongation, alignment, and coupling to form trains across the donor characteristics, procedures and neuropathology. Curr Alzheimer Res somatosensory cortex after experimental diffuse brain injury. J 9(6):724–733 Neuroinflammation 9. doi:10.1186/1742-2094-9-247 16. Bachstetter AD, Rowe RK, Kaneko M, Goulding D, Lifshitz J, Van Eldik LJ 38. Cao T, Thomas TC, Ziebell JM, Pauly JR, Lifshitz J (2012) Morphological and (2013) The p38alpha MAPK regulates microglial responsiveness to diffuse genetic activation of microglia after diffuse traumatic brain injury in the rat. traumatic brain injury. J Neurosci 33(14):6143–6153. doi:10.1523/ Neuroscience 225:65–75. doi:10.1016/j.neuroscience.2012.08.058 JNEUROSCI.5399-12.2013 39. Ackman JB, Siddiqi F, Walikonis RS, LoTurco JJ (2006) Fusion of microglia with 17. Bachstetter AD, Norris CM, Sompol P, Wilcock DM, Goulding D, Neltner JH, pyramidal neurons after retroviral infection. J Neurosci 26(44):11413–11422. St Clair D, Watterson DM, Van Eldik LJ (2012) Early stage drug treatment doi:10.1523/JNEUROSCI.3340-06.2006 that normalizes proinflammatory cytokine production attenuates synaptic 40. Torres-Platas SG, Comeau S, Rachalski A, Dal Bo G, Cruceanu C, Turecki G, dysfunction in a mouse model that exhibits age-dependent progression of Giros B, Mechawar N (2014) Morphometric characterization of microglial Alzheimer's disease-related pathology. J Neurosci 32(30):10201–10210. phenotypes in human cerebral cortex. J Neuroinflammation 11. doi:10.1186/ doi:10.1523/JNEUROSCI.1496-12.2012 1742-2094-11-12 18. Schmider E, Ziegler M, Danay E, Beyer L, Bühner M (2010) Is It Really 41. Beynon SB, Walker FR (2012) Microglial activation in the injured and healthy Robust? Methodology: European Journal of Research Methods for the brain: What are we really tallkine about? Practical and theoretical issues Behavioral and Social Sciences 6(4):147–151. doi:10.1027/1614-2241/a000016 associated with measurement of changes in microglia morphology. 19. Leverenz JB, Agustin CM, Tsuang D, Peskind ER, Edland SD, Nochlin D, Neuroscience 225:162–171. doi:10.1016/j.neuroscience.2012.07.029 DiGiacomo L, Bowen JD, McCormick WC, Teri L, Raskind MA, Kukull WA, 42. Nelson PT, Abner EL, Schmitt FA, Kryscio RJ, Jicha GA, Smith CD, Davis DG, Larson EB (2002) Clinical and neuropathological characteristics of Poduska JW, Patel E, Mendiondo MS, Markesbery WR (2010) Modeling the hippocampal sclerosis - A community-based study. Arch Neurol association between 43 different clinical and pathological variables and the 59(7):1099–1106. doi:10.1001/archneur.59.7.1099 severity of cognitive impairment in a large autopsy cohort of elderly 20. Nelson PT, Schmitt FA, Lin YS, Abner EL, Jicha GA, Patel E, Thomason PC, persons. Brain Pathol 20(1):66–79. doi:10.1111/j.1750-3639.2008.00244.x Neltner JH, Smith CD, Santacruz KS, Sonnen JA, Poon LW, Gearing M, Green RC, Woodard JL, Van Eldik LJ, Kryscio RJ (2011) Hippocampal sclerosis in advanced age: clinical and pathological features. Brain 134:1506–1518. doi:10.1093/brain/awr053

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: May 23, 2015

There are no references for this article.