A novel, rapid, quantitative cell-counting method reveals neuropathological abnormalities
in psychiatric disorders: a postmortem brain study
Yoshitaka Hayashi
Department of Biological Science
Graduate School of Science and Technology
C ontents
Summary
Abbreviations
General Introduction
Table
Chapter I: E stablishment of a
frozen brains
Summary
Introduction
Materials and Methods
Results
Discussion
Figure legends
Figures
Table
novel, rapid and quantitative cell-counting method for
1-2
3
4-8
9-11
unfixed
12
13
14-15
16-19
20-21
22-23
24-26
27-30
31
Chapter I I : A novel, rapid, quantitative cell-counting method reveals reduction in the frontopolar cortex in major depressive disorder
Summary
Introduction
Materials and Methods Results
Discussion Figure legends
Figures Tables
Chapter I I I : Neuropathological similarities and differences between schizoph disorder : a flow cytometric postmortem brain study
Summary
Introduction
oligodendr oglial
32
33
34-35
36-37
38-39
41-42
43
44-46
47-48
renia and bipolar
49
50
51-52
Figures
Tables
General Discussion
Figure legends
Figures
References
Acknowledgment
69-75
76-85
86-94
95
96-97
98-120
121
Summary
Recent histopathological studies of human postmortem brain suggest that relatively similar trait abnormalities are present in schizophrenia, bipolar disorder, and major depressive disorder. However, the precise neuropathological abnormalities characterizing such clinical similarities, and differences, have not yet been fully investigated, mostly due to technical limitations associated with conventional histopathology.
In the present study, I developed a quantitative cell-counting method for frozen unfixed postmortem brains using a flow cytometer. Anisotropic frozen brain tissue was transformed into an isotropic suspension of nuclei and immunostained with nuclear (7-amino-actinomycinD), neuronal (NeuN), and oligodendroglial (olig2) markers. I counted stained nuclei and measured quantitatively their sizes and fluorescence intensities. With this new approach I have successfully surmounted several disadvantages encountered in standard histopathology, including loss of immunoreactivity and tissue shrinkage. Using this method, the frontopolar and inferior temporal cortical gray matter of patients with schizophrenia (frontopolar cortex, n = 10; inferior temporal cortex, n = 11), bipolar disorder (frontopolar cortex, n = 12; inferior temporal cortex, n = 11), and, major depressive disorder (frontopolar cortex, n = 10; inferior temporal cortex, n = 11), as well as that of normal controls (frontopolar cortex, n = 12; inferior temporal cortex, n = 12) were analyzed.
I found significant reduction in the densities of oligodendrocyte lineage (olig2(+)) nuclei
in the frontopolar cortex from patients with bipolar disorder and major depressive disorder,
compared with control subject. Abnormal distributions in the NeuN(+) neuronal nuclei size,
quantifying the densities of total, neuronal, oligodendroglial and other non-neural nuclei. The findings in the frontopolar and inferior temporal cortices from three psychiatric disorders using the FCM measurement indicate that, while similar neuropathological abnormalities are shared by patients with major psychiatric disorders, differences also exist, mainly between the frontopolar cortex of schizophrenia and mood disorder patients, which might at least partially explain the differences observed in the clinical manifestations of these disorders.
2
Abbreviations
7-AAD, 7-amino-actinimycinD; MDD, major depressive disorder; BPD, bipolar disorder; SCH, schizophrenia; PFC, prefrontal cortex; ACC, anterior cingulate cortex; TC, temporal cortex; FPC, frontopolar cortex; BA, Brodmann area; DLPFC, dorsolateral prefrontal cortex; OFC, orbitofrontal cortex; STC, superior temporal cortex; MTC, medial temporal cortex; ITC, inferior temporal cortex;
OL, oligodendrocyte; OPC, oligodendrocyte progenitor cell; GAD-67, glutamic acid;
decarboxylase-67kD; GABA, y -aminobutyric acid; GAP-43, growth-associated protein-43kD;
SNAP-25, synaptosomal associated protein-25kD; GR, glucocorticoid receptors; 3-D,
three-dimensional; FCM, flow cytometer; PMI, postmortem interval; FS, forward scatter; IGF,
insulin growth factor; BDNF, brain derived neurotrophic factor; NRG, neuregulin; DISC1, disrupted
in schizophrenia 1
General Introduction
A hundred years ago, Emil Kraeplin first described the psychiatric syndromes, now called mood disorders, such as major depressive disorder (MDD) and bipolar disorder (BPD), and schizophrenia (SCH). He was convinced that they were organic brain diseases, and his colleague Alzheimer also researched the neuropathological investigation of psychiatric disorders before moving to a fruitful research area of neurodegenerative disorder, now called Alzheimer's disease. However, no specific abnormalities, such as those observed in neurodegenerative disorders, was found in these investigation, then the subject has still continued to fascinate but subsequently exasperate researchers.
From that time, the neuropathology of psychiatric disorders has been called the "graveyard of neuropathologists" (Plum, 1972).
By 1980, the growing evidence of structural brain changes in psychiatric disorders provided by neuroimaging studies had spurred a return to postmortem studies. The neuroimaging findings in psychiatric disorders are partly but not unequivocally corroborated by measurement of brain postmortem (Harrison, 1999). Alterations in neuronal and glial density and/or size in prefrontal (PFC), anterior cingulate (ACC), and temporal (TC) cortices of psychiatric disorders were also reported from this time (Table 1).
The PFC is the anterior part of the frontal lobes, lying in front of motor and premorter areas. The PFC includes the regions of frontopolar (FPC, Brodmann area [BA] 10), dorsolateral PFC (DLPFC, BA9,46), orbitofrontal cortices (OFC, BA11,12,13,47) (Brodmann, 1909; Ramnani and Owen, 2004). The functions of these regions so far revealed are that the DLPFC may be associated with working memory (Miller and Cohen, 2001), that the OFC may be associated with decision-making and hedonic process (Kringelbach, 2005), and that the FPC functions may integrate above functions and overcome the limitation of more posterior prefrontal processes (Koechlin and Hyafil, 2007). The ACC is the frontal part of the cingulate cortex, a part of the limbic system situating in the medial aspect of the cortex, and is associated with emotion, sensory, motor, and
4
cognitive processes (Vogt et al., 1992; Bush et al., 2000). The TC is a region of the cerebral cortex that involves superior (STC), medial (MTC), and inferior (ITC) region. The STC includes primary auditory cortices (Heschl's gyms, BA 41,42) and a language-related area (Wernicke' area, BA22) (Binder et al., 2000). The MTC (BA21) and ITC are involved in several cognitive process (Cabeza and Nyberg, 2000), including language and semantic memory processing (MTC, BA21) (Tranel et al.,
1997; Chao et al., 1999; Cabeza and Nyberg, 2000), visual information (ITC, BA20) (Riches et al., 1991; Ishai et al., 1999; Herath et al., 2001), and multimodal sensory integration (Mesulam, 1998).
In the postmortem brain studies, Benes et al. first reported that reduction of neuronal density in the FPC layer VI and the ACC layer V, and a trend of reduction in glial cell density throughout most layers in both brain region from patient with SCH using Nissl-stained sections (Benes et al., 1986). Five years later, they reported, in Nissl-stained sections from patients with SCH, reduced numbers of small non-pyramidal interneurons in the FPC layers I and II and increased numbers of large pyramidal neurons in the FPC layer V without concomitant total neuronal loss (Benes et al., 1991). Subsequent studies also observed abnormal interneuronal densities or related gene or protein expressions using Nissl-stained section, in-situ hybridization, immunohistochemistry, immunoblotting, and/or microarray analysis in the FPC, DLPFC, OFC, and/or ACC from patients with SCH (Table 1). Furthermore, some of these changes were also observed in BPD and MDD brains (Guidotti et al., 2000; Beasley et al., 2002; Cotter et al., 2002b; Rajkowska et al., 2007) (Table
1).
A study reported reduction in glial cell densities in the ACC from patient with MDD and
white matter of DLPFC using Nissl stained sections and found reduced density of gray matter OLs in layers III and VI in SCH, BPD, and MDD samples relative to control samples (Uranova et al., 2004).
The postmortem brain studies of the TC revealed gray matter volume reduction in the left STC and a trend of reduction in the MTC and ITC in SCH (Akbarian et al., 1993; Highley et al., 1999). Reduction of protein levels of reelin and glutamic acid decarboxylase-67kD (GAD-67) in the y -aminobutyric acid (GABA)-ergic interneurons were observed in the STC of SCH (Impagnatiello et al., 1998). Altered expression of OL lineage-related genes were also observed in the MTC of MDD (Aston et al., 2005). On the other hand, alterations of several neuronal protein expressions were also observed in the ITC. The levels of growth-associated protein-43kD (GAP-43), a membrane phosphoprotein implicated in the initial establishment and reorganization of synaptic connections, were increased, whereas those of synaptophysin and synaptosomal associated protein-25kD (SNAP-25), synaptic vesicle proteins implicated in the information on the pathways of synaptogenesis and synaptic vesicle release, were decreased in the ITC of SCH (Perrone-Bizzozero et al., 1996; Thompson et al., 1998). A report revealed reduction of glucocorticoid receptors (GR) mRNA hybridization signals in layers IV of ITC by using in-situ hybridization from MDD, BPD, SCH and control subjects (Webster et al., 2002). However, there is no neuropathological evidence for quantitative assay of neuronal and/or glial densities in the ITC from psychiatric disorders.
Because stereological techniques are laborious and intrinsically low throughput, taking typically long periods to complete a large study, only few three-dimensional (3-D) morphometric studies have been conducted to examine the cytoarchitectural abnormalities found in these disorders.
Furthermore, most of these investigations used conventional Nissl staining, because of the advantage of ease and familiarity of use and robustness across a range of fixation and processing conditions.
The fact that the Nissl method stains both neurons and glia is often advantageous, but it is also disadvantageous in that small neurons may be misclassified as glia, or vice versa (Gittins and Harrison, 2004).
6
Recently, Herculano-Houzel, Lent, and colleagues developed a novel cell-counting method to quantify total numbers of neuronal and non-neuronal cells in fixed brains by transforming highly anisotropic brain structures into homogeneous, isotropic suspensions of cell nuclei and by counting them on a hemocytometer (Herculano-Houzel and Lent, 2005; Herculano-Houzel et al., 2008). In the present study, I extended this concept to frozen, unfixed brains, and using a flow cytometer (FCM), quantified comprehensively the numbers, size, and fluorescent intensities of 7-AAD(+), a DNA marker, nuclei immunolabeled for the neuronal and oligodendroglial nuclear markers NeuN and olig2, respectively. NeuN is an antibody that recognizes a neuron-specific antigen and selectively stains neuronal nuclei (Mullen et al., 1992), while olig2 is a bHLH transcription factor that regulates key stages of early OL and motor neuron development (Lu et al., 2002; Takebayashi et al., 2002).
The olig2 marker is frequently used as a pan-oligodendroglial cell type marker in adult mammalian brains (Kitada and Rowitch, 2006; Ligon et al., 2006; Kuhlmann et al., 2008).
In the preliminary study, I estimated the number of total (7-AAD(+)) and NeuN(+)nuclei in the cerebral cortex from unfixed, frozen whole rat brain using FCM measurement. These results were comparable with the number of total and neuronal nuclei in the whole rat cerebral cortex estimated by Herculano-Houzel and Lent (2005). Furthermore, I investigated effects of several confounding factors, such as postmortem interval (PMI), gender, and frozen storage days on the numbers of nuclei from unfixed, frozen, rat cerebral cortex.
Then, I, first, applied this FCM method to frozen unfixed postmortem human brains of the
FPC and ITC from patients with MDD and from demographic factor-matched normal control.
FCM measurement might provide insights into the cellular
similarities and differences between psychiatric disorders.
mechanisms underlying the clinical
8
Table 1. Neuropatholog ical findings in the frontal, anterior cin gulate and temporal cortices of
postmortem brains from patients with major depressive disor der, bipolar disorder, and schizoph renia.
Author Diagnostic Brain region Methods Main finding
Prefrontal and anterior cinoulate cortex
Benes et al., 1986 Schizophrenia BA10, BA24 Nissl stain
Decreased neuronal density in layer VI of BA10, layer V of BA24 and a trend to decrease glial densitythroughout most layers of both regions.
Benes et al., 1991 Schizophrenia, Schizoaffective BA10, BA24 Nissl stain
Small interneurons decreased, mainly layer II in BA 10 from both subjects. Small interneuronal decreased in layer II-VI of BA24 from schizoaffective disorders. Pyramidal neurons increased in layer V of BA 10 from both subjects.
Akbarian et al., 1995
Selemon et al., 1995
Beasley and Reynolds. 1997
Ong ur et al., 1998
Schizophrenia
Schizophrenia
Schizophrenia
Major dipressive disorders Bipolar disorders Schizophrenia
BA9
BA9 (left hemisphere) BA10
BA24 (left hemisphere) ISH
Nissl stain IHC
Nissl stain
Decreased densityof GAD67 mRNA(+) neurons in layer II-V.
Increased neuronal densityin layer III-VI.
Decreased densityof parvalbumin(+) neurons in layer III-IV.
Decreased glial densityin major depressive and bipolar disorders but not in schizophrenia. Unaltered neuronal densityin three psychiatric disorders.
Rajkowska et al., 1998 Schizophrenia BA9 (left hemisphere) Nissl stain
Decreased neuronal soma sizes accompanied by increases in the densityof small neurons. In layer III only, a significant reduction in mean neuronal size was associated with a significant decrease in the densityof very large neurons.
Selemon et al., 1998
Honer et al., 1999
Rajkowska et al., 1999
Guidotti et al., 2000
Kozlovsky et al., 2000
Schizophrenia Major depressive disorder Schizophrenia Major depressive disorder Major depressive disorders Bipolar disorders Schizophrenia Major depressive disorder Bipolar disorder Schizophrenia
BA46 (left hemisphere)
BA10
lateral orbitofrontal cortex (BA10-47, left hemisphere)
BA9
BA10
Nissl stain
IB
Nissl stain
IB
IB
Increased neuronal densityin layers II-IV and VI.
Decreased protein level of myelin basic protein in major depressive disorder and schizophrenia.
Decreased neuronal sizes in layer II-III and glial density in layer III-IV.
Decreased levels of reel in and GAD67 proteins in the bipor disorders and schizophrenia but not in major depressive disorders.
Decreased GSK-313 protein !eves in schizophrenia but not in mood disorders.
Mirnics et al., 2000 Schizophrenia BA9 Microarray Decreased expression levels of presinaptic secretory machinery,
gulunatate system, and GABA system related genes.
Beasley et al., 2001 Schizophrenia BA9 IB Decreased GSK-36 protein levels in schizophrenia.(3-catenin and
dishevelled-2 protein levels were not altered.
Cotter et al., 2001
Major dipressive disorders Bipolar disorders Schizophrenia
BA24 Nissl stain
Glial cell density and neuronal size were reduced in layer VI in major depressive disorder. Reduced glial cell densityin layer VI of schizophrenia and unaltered in bipolar disorders.
Hakaket al., 2001 Elderly schizophrenia BA46 Microarray
Alterated expression levels of signal myelination, synaptic plasticy, neuronal development, GABA system, neurotransmission, and signal transduction related genes.
Rajkowska et al., 2001 Bipolar disorders BA9 (left hemisphere) Nissl stain Decreased neuronal (layer III), pyramidal cell (layer III, V), and glial (layer III) densities in bipor disorders.
Vol k et al., 2001 Schizophrenia BA9 ISH Decreased densityof GAD67 mRNA(+) neurons in layer III-V.
Beasley et al., 2002
Major depressive disorders Bipolar disorders Schizophrenia
BA9 IHC
Decreased calbindin(+) neurons in bipolar dosrders (layer II-III) and schizophrenia (layer II-III, V). Decreased parvalbumin (+) neurons in layer III of schizophrenia. No alteration in major depressive disorders.
Cotter et al., 2002a
Major depressive disorders Bipolar disorders Schizophrenia
BA9 Nissl stain
Decreased in glial cell density in layer V and neuronal size in layer VI
in major depressive disorders. Decreased in glial densityin layer V in
schizophrenia while neuronal size was reduced in layers vand VI in
Table 1. continued
Author Diagnostic Brain region Methods Main finding
Selemon et al., 2003 Schizophrenia BA9, BA44 (Broca's area)
(left hemisphere)Nissl stain Increased neuronal density in BA9 but there is no alteration in BA44.
Tkachevet al., 2003 Bipolar disorder
Schizophrenia BA9 q RT-PCR Decreased expression levels of oligodendrocyte-related and myelin-
related genes (e.g. PLP1, MAG, CLDN11, SOX10, MOBP, ERBB3, TF)
Veldic et al., 2004 Schizophrenia BA10 IHC, ISH
NeuN(+) neuronal density was unchanged. Increased DNMT1 mRNA(+) neuronal densityin layer I-IV and decreased REELIN mRNA(+) neuronal densityin layer I-II and IV.
Woo et al., 2004 Schizophrenia
Bipolar disorders BA24 ISH
Decreased density of GAD67 mRNA(+) neurons that co-express N R2A mRNA in layer II and V of schizophrenia and layer II of bipolar disorders.
Cotter et al., 2005
Major dipressive disorders Bipolar disorders Schizophrenia
Orbitofrontal crtex Nissl stain
Reduced neuronal sizes in layer I of bipolar disorders. No evidence for group differences in glial cell size nor for differences in glial or neuronal density.
Hashimoto et al., 2005 Schizophrenia BA9 ISH Decreased levels of GAD67 mRNA. The deficit in GAD67 mRNA was
correlated with decreased levels of BDNF/TrkB mRNAs.
Iwamoto et al., 2005
Miguel-Hidalgo et al., 2005
Rajkowska et al., 2005
Schizophrenia
Major depressive disorders Schizophrenia Major depressive disorders
BA10
BA9 (left hemisphere)
BA47 (left hemisphere)
Methyiation assay q RT-PCR
IHC
Nissl stain
Increased metyiated levels of SOX10 and decreased expression levels of SOX10.
Increased mean size of 200 kD neurofilament protein in pyramidal cells in layer III of schizophrenia.
Decreased pyramidal neuronal densityin layers III and V. Significant negative correlation between age at death and the densityof pyramidal neuron in layer III.
Veldic et al., 2005
Scizophrenia Bipolar disorder (with psychosis)
BA9 ISH, IHC
Increased densityof DNMT1mRNA and protein positive neurons in layer I-II of both psychotic subjects, but not in bipolar disorders without psychosis. Decreased densityof GAD67 mRNA(+) neurons in psychotic disorders. The number of GAD67 mRNA(+) neurons showed negative correlation with those of DN MT1 mRNA(+) neurons.
Cullen et al., 2006 Schizophrenia BA9 Nissl stain Increased neuronal density in left hemisphere of normal control than right
hemisphere. This left-right asymmetry was reversed in schizophrenia.
Raj kowska et al., 2007 Major depressive disorders BA9, BA47 IHC
The densityand size of calbindin(+) neurons was sig nificantly reduced by50% in depression in the BA9 and there was a trend toward reduction in the BA47. In contrast, there was no difference in the densityand size of parvalbumin(+) neurons in BA9 and BA47.
Ruzicka et al., 2007 Schizophrenia BA9 LAM
q RT-PCR
Increased DNMT1 expression in neurons microdisected from layer I, but not in layer V. GAD67 and reelin were underexpressed neurons isolated from layer I and unaltered neurons in layer V.
Veldic et al., 2007 Bipolar disorders
Schizophrenia BA9 ISH
Increased densityof DNMT1 mRNA(+) neurons and decreased density of reelin and GAD67 mRNA(+) neurons but unaltered GAD65 mRNA(+) neuron in both disorders. These alterlations were observed in layer I-II in bipolar disorders and layer I-Ill in schizophrenia.
Vostri kov et al., 2007
Van Otterloo et al., 2009
Kim et al., 2010
Major dipressive disorders Bipolar disorders Schizophrenia Major depressive disorders (>60 years old) Major dipressive disorders Bipolar disorders Schizophrenia
BA9
BA9 (left hemisphere)
BA9
Nissl stain
Nissl stain
Microarray Nissl stain
Decreased perineuronal oligodendrocyte in layer III of all subjects.
No significant differences in laminar densityof pyramidal or non- pyramidal neurons.
Eight hundred eighteen genes were significantlycorrelated with a decrease in the number of perineuronal oligodendrocytes across all subjects. Six hundred genes were sig nificantly correlated with a decrease in densityof calbindin-positive interneurons across all subjects.
Frontal and temporal cortex
Perrone-Bizzozero et al., 1996 Schizophrenia BA9, BA10, BA17, BA20 IB
Increased levels of GAP43 and decreased levels of synaptophysin in BA9, 10, 20 but not in BA17. Protein levels of GFAP was not altered in schizophrenia.
Impag natiel to et al., 1998 Schizophrenia PFC(BA10-46), BA22 qRT-PCR, IHC, IB
Decreased RELN mRNA, increased GABA-A receptors al and a2 mRNA, and decreased RELN(+) neuronal densities in layer I-II in PFC.
Decreased protein levels of RELN and GAD67 in BA22.
Tompson et al., 1998 Schizophrenia BA9, BA10, BA20 IB Increased SNAP-25 protein level in BA9 and BA20, and this protein level
was reversed in BA10.
Temporal cortex
AIQ>arian et al., 1993 Schizophrenia BA21 IHC Decreased NADPH-d(+) neuronal densityin gray matter, while
increased NADPH-d(+) neuronal densityin white matter.
10
Table 1. continued
Author Diagnostic Brain region Methods Main finding
H ig hley et al., 1999
Webster et al., 2002
Sweet et al., 2002
Aston et al., 2005
Beasleyet al., 2005
Chance et al., 2005
Beasleyet al., 2009
Schizophrenia Major dipressiNe disorders Bipolar disorders Schizophrenia Schizophrenia
Major depressive disorders
Major dipressNe disorders Bipolar disorders Schizophrenia Schizophrenia Major dipressi\.e disorders Bipolar disorders Schizophrenia
BA20, BA21, BA22
BA20
BA42
BA21
planum temporale planum temporale planum temporale (white matter)
Brain specimens
ISH
Nissl stain
Microarray
Nissl stain
IHC
Nissl stain
Decreased g ray matter volume in the left BA22 and a decrease trend in the BA20. Unaltered in BA21.
Decreased glucocorticoid receptor mRNA level in layer IV of all psychiatric disorders.
Decreased somal volume of pyramidal neurons. Percent change in somal olume was correlated between BA42 and BA9.
Altered the expression of oligodendrocyte function related genes (e.g.
CNP, MAG, MAL, MOG, MOBP, PMP22, PLLP, PLP1, ASPA, UGT8, ENPP2, EDG2, TF, KLK6, SOX10, OLIG2, ERBB3)
Decreased neuronal sizes in layer III and glial densityin layer VI of bipolar disorders.
Decreased calbindin(+) neuronal densityin layer II.
Decreased g lial density in schizophrenia. No significant differences in bipolar disorders and major depressive disorders.
IHC, immunohistochemistry; IB, immunoblotting; ISH, in-situ hybridization; qRT-PCR, quantitative real-time polymerase chain reaction; LAM, laser-assisted microdissection; SNAP-25, 25 kDa synaptosome-associated protein; GAD, glutamic acid decarboxylase; DNMT, DNA-methyltransferase; NR2A, subunit of N-methyl-D-aspartate (NMDA) receptor complex;
GSK-313, glycogen synthase kinase-3; CNP, 2'3'-cyclic-nucleotide 3'-phosphodieserase; MAG, myelin associated glycoprotein; MAL, myelin and lymphocyte glycoprotein; MOG, myelin oligodendrocyte glycoprotein; MOBP, myelin-associated oligodendrocyte basic protein; PMP22, Peripheral myelin protein 22; PLLP, plasmolipin, a myelin protein; PLP1, lipophilin, primary constituent of myelin; ASPA, aspartoacylase; UGT8, cerebroside synthase; ENPP2, autotaxin, myelin formation; EDG2, oligodendrocyte/myelin maintenance; TF, transferrin; KLK6, kallikrein 6;
SOX10, transcription factor, regulates myelin-related genes; OLIG2, oligodendrocyte lineage
transcription factor; ERBB3, neuregulin receptor tyrosine kinase.
Chapter I
E stablis hment of a novel, rapid, and quantitative cell-counting method for unfixed frozen brains
12
Summary
Stereological techniques that estimate cell numbers must be restricted to well defined structures of isotropic regions and therefore do not apply to anisotropic regions of brain. I developed a novel, fast, quantitative cell-counting method for frozen unfixed postmortem brains using a flow cytometer.
Anisotropic frozen brain tissue was transformed into an isotropic suspension of nuclei and immunostained with nuclear (7-AAD), neuronal (NeuN), and oligodendroglial (olig2) markers. I counted stained nuclei and measured their sizes and fluorescence intensities. Furthermore, using this novel cell-counting method, the effects of postmortem interval, storage days, and gender on the numbers, size, or immunoreactivities of nuclei from unfixed, frozen adult rat cerebral cortex were evaluated. In the adult rat cerebral cortex, I show that it contains —76 million cells, of which 36 million and 15 million are neurons and oligodendroglias, respectively. Furthermore, the postmortem interval, freezing storage days, and gender did not affect these numbers and immunoreactivity.
However, the postmortem interval altered the distribution of NeuN(+) forward scatter (cell size)
values. This novel method provides a simple, rapid cell-counting method, quantifying the densities of
total, neuronal, oligodendroglial and other non-neural nuclei comprehensively. Although the value of
postmortem interval should be required utmost care, the distribution of FS values roughly represents
neuronal subpopulation. The novel method will be a useful tool for future neuropathological
research.
ntroduction
Estimation of neuronal and glial cell numbers is important in the neuropathological research. In the gray matter of adult rodent and human cortex, at least four major glial subtypes have been identified:
astrocytes, oligodendrocytes (OLs), oligodendrocyte progenitor cells (OPCs), and microglia (Nishiyama et al., 2009). Stereological methods have been used to estimate the number of cells, neurons, and glias in discrete brain regions (Andersen et al., 1992; West, 1999). However, stereological techniques are laborious and intrinsically low-throughput, taking typically large terms to complete a large study. Furthermore, most of these investigations have used conventional staining of the Nissl substance with cresyl violet, because of the advantage of ease and familiarity of use, and robustness across a range of fixation and processing conditions. The fact that it stains both neurons and glia is often advantageous, but it is also a disadvantage in that small neurons may be misclassified as glia, or vice versa (Gittins and Harrison, 2004).
In the human postmortem brain studies, there are several confounding factors to be seriously considered, including postmortem interval, freezing storage days, and/or gender. These confounding factors are the serious concerns in the neuropathological investigations of postmortem brain tissue, because, for example, these factors often influence the quantification of mRNA (Burke et al., 1991;
Barton et al., 1993; Harrison et al., 1995; Preece and Cairns, 2003).
Recently, Herculano-Houzel, Lent, and colleagues developed a novel cell-counting method to quantify total numbers of neuronal and non-neuronal cells in fixed brains by transforming highly anisotropic brain structures into homogeneous, isotropic suspensions of cell nuclei and by counting them on a hemocytometer (Herculano-Houzel and Lent, 2005; Herculano-Houzel et al., 2008). In the present study, I extended this concept to frozen, unfixed brains, and using a flow cytometer (FCM), quantified comprehensively the numbers, size, and fluorescent intensities of 7-amino-actinomycinD (7-AAD(+)), a DNA marker, immunolabeled for the neuronal and oligodendroglial nuclear markers NeuN and olig2, respectively. NeuN is an antibody that recognizes a neuron-specific antigen and
14
selectively stains neuronal nuclei (Mullen et al., 1992), while olig2 is a bHLH transcription factor that regulates key stages of early OL and motor neuron development (Lu et al., 2002; Takebayashi et al., 2002). The olig2 marker is frequently used as a pan-oligodendroglial cell type marker in adult mammalian brains (Kitada and Rowitch, 2006; Ligon et al., 2006; Kuhlmann et al., 2008).
The FCM measurements enabled me to determine the number of total (7-AAD(+)), NeuN(+),
olig2(+), and NeuN(-)/olig2(-) nuclei in the whole rat cerebral cortex, some of which were
comparable with the number of total and neuronal nuclei in the whole rat cerebral cortex estimated
by Herculano-Houzel and Lent (2005). Furthermore, the FCM measurement also enabled me to
investigate the effects of the postmortem interval (PMI), gender, or frozen storage days on the
numbers and size of each nucleus.
Materials and Methods
All experimental protocols were approved by the Ethics Committee of the Tokyo Metropolitan Institute of Medical Science.
Animals
Wistar rats (Oriental Yeast Co., Ltd., Tokyo, Japan) were maintained in a temperature-controlled room (-23°C) with a 12 h/12 h light/dark cycle, in accordance with the guidelines for the Animal Use and Care Committee of the Tokyo Metropolitan Institute of Medical science. Experiments were performed during the light period. Eight-week-old rats (male, n=8; female, n=4) were killed with a lethal dose of anesthesia (pentobarbital sodium), and then perfused with phosphate-buffered saline (PBS) containing 50 IU/ml heparin. The brains were then removed from the skull, and the cerebral cortex was dissected out as described by Herculano-Houzel and Lent (2005). The brains were immediately frozen and stored at -80°C until measurements were carried out. The frozen brains were used within a month, except for a group of brains from 8-week-old rats (n=4), which were stored for 385 days in order to investigate the effects of frozen storage days.
For our investigation on the effects of postmortem interval (PMI), rats (8-week-old, male, n=4) were killed by either a lethal dose of anesthesia or by cervical dislocation. The animals were then kept at 4°C for 0, 24, or 48 hours without perfusion. At each PMI, brains were removed from the skull, the cerebral cortex was dissected out as described above, and the tissue was immediately frozen and stored at -80°C until measurements were carried out. These samples differed from the perfused samples in that they contained blood cells.
Nuclei isolation
The rat brain samples were dissolved and wet weights were measured. A crude suspension of nuclei
was obtained through mechanical homogenization of the brain tissue in a 10-fold volume of lysis
16
buffer (V/W) (0.32 M sucrose, 5 mM CaC12, 3 mM magnesium acetate, 0.1 mM EDTA, 10 mM Tris-HC1 [pH 8.0], 0.1% Triton X-100) (Spalding et al., 2005). Homogenization was performed with a set of 1.5 ml Eppendorf tubes and a plastic pestle by grinding for about 1 minute. This isolation method dissolved the cytoplasmic membrane and organelles but left the nuclear membrane intact.
The homogenized brain tissues were immediately centrifuged (10 min at 1000x g) and the resulting pellets containing nuclei and debris were collected. To detect all nuclei, I used 7-aminoactinomycin D (7-AAD), a fluorescent DNA marker whose emission has a very large stroke shift, with the maximum occurring in the deep red range of 647 nm. The pellets were resuspended in PBS containing 10 µ1/m1 7-AAD (BD Pharmingen, Franklin Lakes, NJ) and 10 gl/m1 RNase A (50 µg/ml) (Nacalai Tesque, Kyoto, Japan). The final volume of the nuclear suspension was adjusted to 10 mg tissue weight per 1 ml suspension solution.
Nuclei counting
Flow-Count® (Beckman Coulter, Fullerton, CA) was used to count the number of nuclei.
Flow-Count® contains control beads of identical bead size that have high fluorescence emissions
across all ranges of the flow cytometer; 1 µl of Flow-Count® contains a finite number of beads. An
aliquot of 7-AAD-stained nuclei was mixed with an equal amount of Flow-Count® prior to counting
the nuclei by FCM (Epix XL, Beckman Coulter, Fullerton, CA). Nuclei counting continued until the
flow cytometer had counted 1000 control beads. The exact number of nuclei was determined by
averaging the numbers obtained from at least three independent measurements. In preliminary
After the nuclei counts, I centrifuged the nuclear suspension at 1000g for 10 min at room temperature (RT). The resulting pellet was resuspended in 100 µl of blocking solution (3% BSA, 2%
skim milk in PBS) and incubated for 30 min at RT. Subsequently, the blocking solution was centrifuged, and the resulting pellet was resuspended in 100 µl of blocking solution containing mouse anti-NeuN IgG (1:1000; Chemicon, Temecula, CA) and rabbit anti-olig2 IgG (1:2000;
Chemicon) or isotype IgG controls (anti-mouse IgG from BD Pharmingen; anti-rabbit IgG from Southern Biotech, Birmingham, AL) and incubated overnight at 4°C. I chose NeuN and olig2 as neuronal and oligodendroglial nuclear markers, respectively, partly because I could not find any established and appropriate nuclear markers for other neural cells, such as astrocytes, microglia, and endothelial cells.
The next day, 1 ml of PBS/0.05% Tween20 was added to the suspension, and the whole was centrifuged. The resulting pellet was resuspended and incubated in blocking solution containing anti-mouse PE (Molecular Probes, Burlington, Canada) and anti-rabbit Alexa488 (Molecular Probes) for 1 h at RT. Thereafter, stained nuclei were suspended in PBS containing 10 gl/m1 7-AAD and 50 µg/ml RNase A. The results of the FCM analyses are presented as percentages of NeuN(+), olig2(+), and NeuN(-)/olig2(-) nuclei in 20,000 7-AAD(+) nuclei. Each absolute number of nuclei was calculated by multiplying the number of 7-AAD(+) nuclei by the percentage of NeuN(+), olig2(+), and NeuN(-)/olig2(-) nuclei.
Statistical Analysis
Statistical analyses were carried out using an unpaired Student's t test for two-groups and a one-way analysis of variance (ANOVA) followed by a Dunnett's post hoc test for more than two groups at the 5% significance level with GraphPad PRISMTM (Graph Pad software, Inc., San Diego, CA) and SPSSTM (SPSS, Inc., Chicago, IL) software.
For a comparison of the FS distribution of NeuN(+) nuclei, I chose two FS ranges
18
(FS 100-FS 199 and FS250-FS349) and performed an unpaired Student's t test for two-groups and a
one-way analysis of variance (ANOVA) followed by a Dunnett's post hoc test for more than two
groups at the 5% significance level.
Results
FCM measurement of rat cortical cell nuclei
Using an adult rat cerebral cortex, I first examined whether the isotropic fractionator method established by Herculano-Houzel and Lent (2005) could be used to measure nuclei obtained from frozen unfixed brain tissue and if counting could be accomplished using FCM instead of by visual counting with a hemocytometer. Although the wet weights of the rat cerebral cortices differed considerably across the two studies (1054 mg vs. 770 mg) — perhaps because the tissue used in this study was unfixed, while the tissue used by Herculano-Houzel and Lent (2005) was fixed--- the number of total (7-AAD(+)) and neuronal (NeuN(+)) nuclei counts in the entire adult rat cerebral cortex proved comparable in both studies (Table 2). In our FCM measurements (Figure 1), the total number of cells in rat cerebral cortex was 75.92 million, 47% of which (35.90 million) were NeuN(+) (Table 2). The co-efficient of variation was <0.13 for the number of 7-AAD(+) nuclei and
<0.11 for that of NeuN(+) nuclei. The number of 7-AAD(+) nuclei was also counted on a hemocytometer (77.04 ± 9.82 million cells in the whole cerebral cortex of adult rat [n = 4]) and was found to be almost equivalent to that counted by FCM (Table 2) and that reported by Herculano-Houzel and Lent (2005). These findings demonstrated that using FCM with frozen unfixed brain homogenates is a reliable way to estimate the numbers of cells in the brain.
At the same time, the flow cytometer also counted the numbers of olig2(+) nuclei, which most likely represented oligodendroglial cells (OLs and OPCs) in the adult cerebral cortex. Notably, olig2(+) and NeuN(+) populations were mutually exclusive (Figure 1 G), whereas olig2(+) populations often exhibited two fluorescence peaks: olig2Weak(+) and olig2strong(+) populations (Figure 1H). Since olig2 appears to be expressed weakly in mature OLs and strongly in the OPCs of adult mouse and human brains (Kitada and Rowitch, 2006; Kuhlmann et al., 2008), the olig2weak(+) and 011g2strong(+) populations roughly represent mature OLs and OPCs, respectively.
Forward scatter (FS) values generally represent nuclear sizes. NeuN(+) nuclei were small to
20
large, while olig2(+) nuclei were mostly small (Figure 1C). Since a high degree of correlation exists between nucleus size and perikaryon volume in neocortical neurons (Stark et al., 2007), small and large NeuN(+) nuclei roughly represent interneurons and pyramidal neurons, respectively. However, one should be cautious about interpreting the FS values of large NeuN(+) nuclei, since under microscopic observation the areas of these cells generally appear larger than that of 7-AAD(+) nuclei (Figure 2).
E ffects of PMI, gender, and frozen storage days on the number of nuclei
I then investigated the effects of PMI, gender, and frozen storage days on the number of nuclei in the rat cerebral cortex. The numbers of nuclei were not affected by confounding factors such as PMI (0, 24, or 48 h, one-way ANOVA, 7-AAD(+), F(2,9)=1,17, P=0.353; NeuN(+), F(2,9)=1,49, P=0.275;
olig2(+), F(2,9)=0.169, P =0.848; NeuN(-)/olig2(-), F(2,9)=0.56, P =0.590), gender (unpaired t test, 7-AAD(+), P =0.175; NeuN(+), P =0.516, olig2(+), P =0.164, NeuN(-)/olig2(-), P =0.542), or frozen storage days (20 vs. 385 days at -80 °C, unpaired t test, 7-AAD(+), P =0.989; NeuN(+), P =0.809, olig2(+), P =0.124, NeuN(-)/olig2(-), P =0.311) (Figure 3).
I also studied the effects of these confounding factors on the FS values of NeuN(+) nuclei
using rats (Figure 4). Interestingly, PMIs (0, 24, or 48 h) significantly changed the distribution of
NeuN(+) nuclei, shifting the peaks toward the larger (right) direction, depending on the PMIs,
without changing the peak heights. This resulted in a PMI-dependent increase in the densities of
larger NeuN(+) nuclei in the rat cerebral cortex (Figure 4A), indicating that caution must be required
Discussion
In this chapter, I introduced a novel quantitative brain-cell-counting method for frozen unfixed brains by using FCM. This method enables us to determine the numbers, sizes, and fluorescent intensities of total (7-AAD(+)), neuronal (NeuN(+)), oligodendroglial (olig2(+)) or other neural cell nuclei (NeuN(-)/olig2(-)) in the frozen unfixed brains. Furthermore, the effect of some confounding factors on the number and FS distribution of nuclei were studied.
Stereological techniques are useful for determining densities of structural components in discrete, isotropic brain regions, such as subcortical nuclei. To minimize the problems of investigation of anisotropic structures, such as gray matter of the cortex, a particular stereological technique proposes to parcel the structure of interest into small volumes that are less heterogeneous and to more accurately measure by microscopy (West, 1999). A cell-counting method made by Herculano-Houzel and Lent (2005) takes this idea by turning brain tissue isotropic, or completely fractionated, through dissociation that preserves only the cell nuclei from fixed brains. In this chapter, I further extended this concept to frozen, unfixed brains, and using a flow cytometer (FCM), quantified comprehensively the numbers, size, and fluorescent intensities of 7-AAD(+) nuclei immunolabeled for the neuronal and oligodendroglial nuclear markers NeuN and olig2, respectively.
The FCM method is fast and easy, does not require specific stereological techniques, and can be applied to the determination of total, neuronal, oligodendroglial, and other non-neural cell numbers in the study of development, adult neurogenesis, and neuropathology of unfixed, frozen rodent brains.
I obtained the evidences that the number of total, neuronal, oligodendroglial, and other non-neural nuclei were not affected several confounding factors, such as PMI, freezing storage days, and gender. These results suggest that those confounding factors do not affect the stabilities of nuclear envelope and immunoreactivity to nuclear markers. Although I found no effects of freezing storage days and gender on the distribution of NeuN(+) nuclei, PMI was found to affect the
22
distribution of FS values of NeuN(+) nuclei significantly (Figure 4), but not other cell nuclei (data not shown). These findings suggest that studies on the FS values from different PMI specimens, especially those of large NeuN(+) nuclei may be inaccurate.
In this chapter, I established a novel, quantitative cell-counting method using FCM. The method could accurately estimate the number of total (7-AAD(+)), neuronal (NeuN(+)), oligodendroglial (olig2(+)), and other non-neural (NeuN(-)/olig2(-)) nuclei from the unfixed frozen rat cerebral cortex.
Furthermore, I obtained the evidence that the number of nuclei from rat cerebral cortices were
unaffected by the PMI, freezing storage days, and gender. However, PMI significantly changed the
distribution of NeuN(+) nuclei. For that reason, an analyses of large NeuN(+) nuclei size from
different PMI specimens may be controversial.
F igure legends
F igure 1. FCM measurement results from an isotropic suspension of nuclei from frozen rat cerebral cortex. (A) 7-AAD(+) nuclei (blue) and cellular debris (gray). The 7-AAD(+) nuclei generally included two populations: a large population (arrow), probably representing nuclei at the GO/G1 phase of the cell cycle, and a much smaller population (arrowhead), representing nuclei at the G2/M phase of the cell cycle or nuclear doublets. (B-G) NeuN(+) (red) and olig2(+) nuclei (green). (B, D, and F) Negative controls stained with control IgG. (C) NeuN(+) nuclei (red) were detected in a population with higher Qdot565 fluorescence intensities. (E) Olig2(+) nuclei (green) were detected in a population with higher Alexa488 fluorescence intensities. (G) NeuN(+) (red), olig2(+) (green), and NeuN(-)/olig2(-) nuclei (blue) were detected by FCM as independent populations.
NeuN(+)/olig2(+) nuclei were rarely observed in this plot, suggesting that NeuN(+) and olig2(+) populations are mutually exclusive. (H) A histogram of the olig2(+) population. Note that, in E, G, and H, the olig2(+) population could be divided into two populations based on fluorescence intensities: olig2strong(+) and olig2Weak(+) These two populations most likely represent oligodendrocyte progenitor cells and mature oligodendrocytes, respectively.
F igure 2. Fluorescence images of rat nuclei. (A-C) 7-AAD(+) (red, A); NeuN(+) (yellow, B); and olig2(+) (green, C) nuclei. NeuN(+) nuclei were either large or small (B), while olig2(+) nuclei were small (C). Note that the large NeuN(+) areas appeared to be larger than the 7-AAD(+) areas, while the small NeuN(+) areas and the small olig2(+) areas appeared to be almost the same as the 7-AAD(+) areas. (D) Merged image of images shown in panels A-C. NeuN or olig2 colocalized with 7-AAD(+) nuclei. Scale bar, 50 gm.
Figure 3. The effect of postmortem interval (PMI), gender, and freezing storage days on the numbers of nuclei in the whole cerebral cortex of an adult rat. (A-C) The effect of PMI, gender, and freezing
24
storage days on the numbers of total (blue), NeuN(+) (red), olig2(+) (green), and NeuN(-)/olig2(-) (purple) nuclei in whole cerebral cortex of adult rats. (A) The brains were dissected out 0 (n=4), 24 (n=4), and 48 (n=4) hours after the rats were sacrificed, and then processed for FCM measurements.
The number of nuclei was not significantly different across the three PMI samples (one-way ANOVA, total, F(2,9)=1,17, P=0.353; NeuN(+), F(2,9)=1,49, P=0.275; olig2(+), F(2,9)=0.169, P=0.848;
NeuN(-)/olig2(-), F(2,9)=0.56, P =0.590). (B) Male (n=4) and female (n=4) rats were perfused, and the brains were dissected out and then processed for FCM measurements. The number of nuclei was not significantly different across genders (unpaired t test, total, P=0.175; NeuN(+), P=0.516, olig2(+), P =0.164, NeuN(-)/olig2(-), P =0.542). (C) Eight-week-old male rats were perfused, and the brains were dissected out, stored in a deep freezer for either 20 days (n=4) or 385 days (n=4), and then processed for FCM measurements. The number of nuclei was not significantly different across the two groups (unpaired t test, total, P =0.989; NeuN(+), P =0.809, olig2(+), P =0.124, NeuN(-)/olig2(-), P =0.311). Note that the number of NeuN(-)/olig2(-) nuclei in the non-perfused samples (A) was greater than that in the perfused samples (B and C), probably because the former samples contained blood cells.
Figure 4. Effects of confounding factors on the FS distribution of NeuN(+) nuclei from the whole
cerebral cortex of an adult rat. (A) Effect of PMI. Brains were dissected out at 0 (blue), 24 (red), and
48 (green) hours after the rats were sacrificed. With increasing PMIs, the FS distribution peaks of the
NeuN(+) nuclei shifted to larger FS values. No significant difference was found in small NeuN(+)
nucleiat20(blue)or385days(red)(unpairedt‑test,FS100‑199,t(6)=‑0.265,P=0.800;FS250‑349, t(6)=0.498,P=0.636).
26
F igure 1
A
103 102 0
101 ti
100
1111110411,0
•••
0
-44
1023 FS
B
103 102 0 oi
100 00,
0 FS 1023
C 103
z 102 Z101
100
D
103
00 017 100
0
F 103--
102=
0)101-a 100
FS
1023
E
0
1031 102
N -
o 100,
FS
1023
0
G103
FS
1023
100 101 IgG
H
36
E C a) U
103 103
102 Z
W)101
100
100 101
olig2 102
100 101
olig2
102 103
103
Figure2
28
Figure3
A
rn
~E
c oo
E° x Z "
100 75 50 25
0 Ohr 24hr 48hr
-^-Total NeuN(+)
• olig2(+)
NeuN(-)/olig2(-)
PMI
B Vg VE c~
4— O L
~`VL E~ x Z"
100 75 50 25
0 Male Female
C 13)~
E c~
o0
ft
100 75 50 25
0 20d 385d
Gender Storage days
Figure4
A
0 0 5 1 5 0 5 0 5 (∠ (U 7 ﹁ 5 (∠
﹂ ■ ー ﹂ ■ ー
一90コεO﹂ΦΩ∈コZ0
'釜 :
PMIOH PMI24H PM148H
B 0 5 1
.・㌃韮毛¶‑・︒義 〜 .ρ 3 を
ノ︒5 0 5 0 5 0 ∩ ∠ 0 7 ﹁ 5 ( ∠ ﹂■ 1 1 一Φ 口O コ ε O ﹂ Φ Ω ∈ コ Z
■Male Female
.L
1025
∩ V 5 ( U 5 (U 5 F O 2 0 7 ・ 5 ( ∠
﹂■1∠■1﹂■l
C一Φ一〇コεO﹂2∈コZ
