A Novel In Vitro Model to Study Pericytes in the
Neurovascular Unit of the Developing Cortex
Christoph M. Zehendner*, Hannah E. Wedler¶, Heiko J. Luhmann
Institute of Physiology and Pathophysiology, University Medical Center of the Johannes Gutenberg-University, Mainz, Germany
Abstract
Cortical function is impaired in various disorders of the central nervous system including Alzheimer’s disease, autism
and schizophrenia. Some of these disorders are speculated to be associated with insults in early brain development.
Pericytes have been shown to regulate neurovascular integrity in development, health and disease. Hence, precisely
controlled mechanisms must have evolved in evolution to operate pericyte proliferation, repair and cell fate within the
neurovascular unit (NVU). It is well established that pericyte deficiency leads to NVU injury resulting in cognitive
decline and neuroinflammation in cortical layers. However, little is known about the role of pericytes in
pathophysiological processes of the developing cortex. Here we introduce an in vitro model that enables to precisely
study pericytes in the immature cortex and show that moderate inflammation and hypoxia result in caspase-3
mediated pericyte loss. Using heterozygous EYFP-NG2 mouse mutants we performed live imaging of pericytes for
several days in vitro. In addition we show that pericytes maintain their capacity to proliferate which may allow cell-
based therapies like reprogramming of pericytes into induced neuronal cells in the presented approach.
Citation: Zehendner CM, Wedler HE, Luhmann HJ (2013) A Novel In Vitro Model to Study Pericytes in the Neurovascular Unit of the Developing Cortex.
PLoS ONE 8(11): e81637. doi:10.1371/journal.pone.0081637
Editor: Edward F Plow, Lerner Research Institute, United States of America
Received August 30, 2013; Accepted October 14, 2013; Published November 21, 2013
Copyright: © 2013 Zehendner et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: CMZ was supported by a Stage 1 grant of the University Medical Center of the Johannes Gutenberg-University of Mainz. This work was
supported by DFG grants to HJL. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the
manuscript.
Competing interests: The authors have declared that no competing interests exist.
* E-mail: Zehendner@uni-mainz.de
¶ This work contains the M.D. thesis of Hanna E. Wedler
Introduction function. However, little is known about the role of pericytes in
the immature cortex during pathological conditions. This is
Proper cortical development and function requires intact partly due to a lack of appropriate in vitro models that allow the
neurovascular coupling [1] and the intact brain endothelial analysis of pericyte-associated pathophysiology under
barrier separating blood from the central nervous system precisely controlled in vitro conditions over prolonged time
(CNS) [1–3]. More than 70% of the cerebrovasculature in the periods. It is speculated that infections and inflammatory
rodent CNS is covered by pericytes [4]. This high percentage of processes during early brain development are implicated in a
microvascular pericyte coverage reflects their pivotal role for a variety of neurological and psychiatric disorders including
proper formation of the blood-brain barrier (BBB), a highly schizophrenia or autism [6–8]. The goal of the present study
organized vascular network within the CNS, which separates was to establish an in vitro model that allows to study pericytes
the neuronal parenchyma from peripheral circulation, and in the developing cortex with a preserved neuronal network and
modulates its supply with nutrients. Besides pericytes various to elucidate the impact of inflammation and hypoxia on
BBB transporters, tight junction proteins and cellular pericytes.
interactions including astrocytes [5], glia and neurons within the Here we present a novel approach to study pericytes for
NVU orchestrate this complex task. Disturbances of this days in cortical organotypic slice cultures (COSC) from
interplay significantly impair neurovascular integrity in the adult newborn postnatal day 3-4 (P3-P4) mice with various methods
as well as developing brain and result in cognitive decline[2]. including live cell imaging and electrophysiological recordings.
Bell et al. have recently demonstrated that pericyte loss impairs Since the developmental stage of the CNS of newborn rodents
learning capability, results in neurovascular impairment and at P3-P4 corresponds to that of preterm human babies at the
leads to accumulation of neurotoxic substances in the cortex of age of postconceptional month 6 [9,10], our clinically relevant
PDGFRb +/- mice starting at 1 month of age [4]. This finding in vitro model allows to elucidate pericyte-associated cellular
points towards a crucial role of pericytes for proper cortical mechanisms in the developing cortex and to unravel processes
PLOS ONE | www.plosone.org 1 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
of repair within the developing brain following e.g. inflammation following day in ice cold acetone (about 10 minutes) for
or ischemia. It may also be a useful approach to assess cell- immunohistochemistry.
based therapies like reprogramming of pericytes into induced
neuronal cells. Cardiac perfusion
C57/BL6 or heterozygous EYFP-NG2[15] mice were
Materials and Methods anesthetized by intraperitoneal (i.p.) injection of ketamine (120
mg/kg bodyweight) and xylazine (16 mg per KG bodyweight).
Ethics statement Body temperature was kept at around 37°C with temperature
All experiments were approved by the ethical committee of controlled heating pads until the stadium of surgical tolerance
the “Landesungersuchungsamt Rheinland-Pfalz” and the according to Guedel[16] was reached. This was assessed by
authority “Landesuntersuchungsamt Rheinland-Pfalz”, protocol absence of a pain reflex upon toe pinch. The thorax was
number: “Aktenzeichen 23 177 – 07 A12-1-004 and 23 177 – carefully opened with a microsurgical pair of scissors and
07 A12-1-005”. Principles of laboratory animal care (European animals were cardially perfused via the left ventricle with icecold Ringer solution containing heparin (1 IU/ml). After removal
laws 86/609/EEC, national laws and NIH publication No. 86-23, of blood the perfusion solution was switched to 2% PFA and
revised 1985) were followed. All efforts were undertaken to perfusion was continued for another 10 minutes. Brains were
minimize the number of animals used and their suffering. collected and post-fixed in 2 % PFA for 2 hours at 4°C. 100 µm
thick coronal sections were cut with a vibratome and used for
Preparation of COSC immunohistochemical processing. For some analyses brains
Cortical organotypic slice cultures were prepared as were collected after blood was removed by saline perfusion
described in detail before with slight modifications[11–13]. All and snap frozen in tissue tek with the help of liquid nitrogen.
efforts were made to minimize the number of animals used and Here, 20 µm thick coronal sections were cut with a cryotome
their suffering. All experimental manipulations were carried out and fixed in ice cold acetone for 4-5 minutes.
according to the European and national laws (86/609/EEC) on
animal handling. In brief C57/BL6 or heterozygous EYFP-NG2 Live Cell Imaging of pericytes
knockin P3/4 mouse pups were rapidly decapitated. The head P3/4 heterozygous EYFP-NG2 mouse pups were deeply
was disinfected with a drop of ethanol 70% and the brain was anesthetized by i.p. injection of ketamine (120 mg/kg
quickly removed and transferred into 4°C cold medium. bodyweight) and xylazine (16 mg per KG bodyweight). After
Beneath a benchtop microscope under laminar flow bulbi and surgical tolerance stadium was reached (tested as documented
the cerebellum were dissected. Hemispheres were carefully above) 50–100 µl tomato lectin were injected into the left
separated and meninges were removed with forceps. ventricle and after 2 minutes animals were perfused with ice
Afterwards hemispheres were cut into 350 µm thick coronal cold heparinized (1 IU/ml) Ringer solution. Thereafter brains
slices using a chopper. Cortices were transferred onto Millicell were collected for COSC procedure as afore (see above)
membrane filters (Merck Chemicals, Schwalbach, Germany) documented. COSC were examined with an upright confocal
that were pre-equilibrated with medium over night at 37°C, 5 % spinning disk system at 37°C in colorless Hanks balanced
CO , humidified atmosphere. Medium was exchanged 1 day saline solution maintaining 1 mmol/l magnesiumchloride, 22
after preparation, thereafter every 2-3 days. Culture medium mmol/l calciumchloride and 6-9 mg/ml glucose.
consisted of 50% MEM HEPES GlutaMax, 25% heat
inactivated horse serum, 25% Hanks balanced salt solution Immunohistochemistry
supplemented with 1 mmol/l magnesiumchloride, 2 mmol/l Immunohistochemical stainings were carried out according to
calciumchloride. Glucose was added to a total concentration of standard procedures described in detail elsewhere[11,17].
6-9 mg/ml, pH was adjusted to 7.2. We would like to stress the Tables 1 and 2 give an overview on used antibodies, dilutions
fact that pH was most important for the preparation and great and purposes of the staining. In brief fixed probes were washed
care should be taken to make sure the pH is correctly adjusted with 0.01 mol/l PBS. Subsequently tissue was blocked with 7%
to 7.2. For culture purposes Millicell filter membranes were pre- normal donkey serum (017-000-121, Dianova, Hamburg) and
incubated in 6 Well plates containing 1 ml medium per well permeated with 0.3% triton in PBS 0.01 mol/l for two hours at
over night. COSC were carefully examined with light room temperature. Only for BrdU staining samples were then
microscopy for intact morphology ahead of experimental incubated for 90 minutes at room temperature with 2 mol/l HCl.
manipulation. Only COSC that displayed an intact morphology To verify the specific binding of the secondary anti-mouse
were used for experimental procedures. antibody (used in Claudin-5 and NeuN stains) mouse IgG wereblocked using a Fab (1:20, 2 hours in PBS 0.01 mol/l,
AffiniPure Fab Fragment Donkey Anti-Mouse IgG (H+L),
5-bromo-2'deoxyuridine (BrdU) application Jackson, Dianova) blocking technique. Primary antibodies were
BrdU was applied similarly as documented elsewhere[14]. In incubated in 2% bovine serum albumin (001-000-161 Diana,
brief COSC were exposed towards medium containing 10 Hamburg) containing 0.05% azide and 0.1% triton in PBS 0.01
µmol/l BrdU for 3 hours on the third day in vitro. Subsequently mol/l (overnight, room temperature). After incubation with
COSC filter membranes were washed gently once and medium primary antibodies probes were washed with PBS 0.01 mol/l
was switched to BrdU-free medium. COSC were fixed the and incubated with secondary antibodies and DAPI in 2%
PLOS ONE | www.plosone.org 2 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Table 1. Primary antibodies and stains. Table 2. Secondary antibodies.
Antibodies and stains Application Secondary antibodies related primary antibodies
endothelial marker, transmembraneous Cy2 (A50-201C, Bethyl Lab,
PDGFR beta (Figure 1 C-F, Movie S1)
anti-claudin 5 (Life technologies tight junction linker protein of the BBB Biomol), 1:200
#35-2500), 1:50 and other endothelial cells e.g. within Cy3 (705-165-147, Jackson, PDGFR beta (Figures 1 G, 2; 3; 4 C; 5 B; 6 A,
gut, kidneys Dianova), 1:200 E, Movies S3, S4, S5)
anti cleaved caspase-3 (Signaling DyLight 488 (711-485-152,
Marker of apoptosis
Technology ASP 175, #9669), 1:200 Jackson, Dianova), 1:100 OR BrdU (Figure 5 A, Movies S5, S6); Ki67
anti-GFAP (Dakocytomation, Z 0334), DyLight 488 produced in donkey (Figure 5 B); cleaved caspase-3 (Figure 7 A),
Marker of astrocytes
1:200 (A120-208 D, Biomol), 1:100 – nitrotyrosine (Figure 7 D, E)
anti-NeuN (Millipore, MAB377), 1:200 Neuronal marker 1:200
anti-PDGFR beta (Neuromics, GT DyLight 488 (A90-337D2, claudin 5 (Figures 2 A, 6 B, Movies S3, S4),
Pericyte marker
15065), 1:200 Biomol), 1:200 NeuN (Figure 2 B)
anti-Ki67 (ab 15580 Abcam), 1:200 Cell proliferation marker Desmin, NG2 (Figures 1 C, F, 4 E, Movies S1,
Alexa Fluor 647 (711-605-152,
Staining against the nucleoside S2), GFAP (Figure 2 A, Movie S2), pan-
anti-BrdU, (347580, Becton Dickinson) Jackson, Dianova), 1:200
analogue of thymidine (BrdU), an Laminin (Figures 3 A, B, Movie S4)
1:100
indicator of DNA replication Cy3 (715-165-151, Jackson,
Claudin 5 (Figures 1 B, E; 3 E)
Tomato lectin, TexasRed (Vector Dianova), 1:200
Endothelial marker
Laboratories) DyLight 549 (712-505-153,
CD105 (Figure 1 D, Movie S1)
DAPI (Sigma, 32670) 0.5 µg/ml Cell nuclei staining Jackson, Dianova), 1:200
anti-Desmin (4024 Cell Signaling), 1:100 Pericyte marker doi: 10.1371/journal.pone.0081637.t002
anti-pan-Laminin (ab7463, Abcam)
Marker of basement membrane
1:1000 Microsystems), Metamorph (Molecular Devices Corp.,
Marker of oxidative stress induced by
Anti-Nitrotyrosine (Merck Millipore, Downington, CA, USA), NIH Fiji is just ImageJ and IMARIS
reaction of peroxynitrite with the amino
06-284), 1:200 imaging software (Bitplane, Switzerland). Caspase-3 positive
acid tyrosine pericytes were identified by visual identification of PDGFR beta
doi: 10.1371/journal.pone.0081637.t001 positive pericytes and cleaved caspase-3 co-localization. The
ratio of pericytes positive for caspase-3 related to all pericytes
bovine serum albumin with 0.05% azide for another 2 hours at was calculated within randomly chosen fields of view (RFV) of
room temperature. After a final washing step in 0.01 mol/l PBS confocal images in COSC in layers II-IV. Images were obtained
probes were embedded in Fluoromount. The sensitivity of the with a Leica SP5 confocal. The ratio of groups treated with IL-1
nitrotyrosine antibody was confirmed by positive controls with beta, hypoxia or both stimuli was set in relation to the mean
peroxynitrite as recommended by the manufacturer. ratio of respective control groups. Relative changes in fold
change are presented. For this purpose 7-41 COSC
Elecytrophysiological Recordings in Multi-Electrode preparations from 6-28 animals (up to 2812 pericytes) were
Arrays analyzed for each condition. Microvascular pericyte coverage
To evaluate spontaneous neuronal network activity of COSC was determined by measuring total vessel length with the Fiji
after five days in vitro, MEA recordings were performed. COSC measurement tool in pixels similar as described before[11].
on Millicell filter membranes were cut out with a small scalpel Total pericyte length was related to vessel length and relative
and transferred on an 8 x 8 MEA chip (multichannel systems, microvascular pericyte coverage was obtained. For evaluation
Reutlingen, Germany) consisting of 60 electrodes with a of caspase-3 inhibition on pericytes loss, pericytes in RFV of
diameter of 30 µm and an interelectrode distance of 200 µm. cortical layers II-IV were identified by DAPI, pan-laminin and
To avoid dislocation of COSC, a small weight was put on top of PDGFR beta co-staining and the total number of pericytes per
the slices. Prior to use the chip was heated up to a temperature volume was determined via confocal z-stacks. Here the
of 37°C. COSC on MEA chips were kept at 20% O2 and 5% minimum spatial resolution was: 1.5 µm X * 1.5 µm Y * 2 µm Z
CO2 in a humidified atmosphere. Recordings were initiated 10 voxel size. Values were related to control groups and
minutes after placing COSC on MEA chips and lasted 30 differences in fold change are displayed. At least 3 COSC per
minutes. MC-Rack software version 4.1.1 (multichannel group were analyzed.
system, Reutlingen, Germany) was used for acquisition of the If not stated otherwise all images shown in this manuscript
data. Sigma Plot version 7.0 (Systrat Software, Erkrath, are confocal images. 3 dimensional reconstructions of z-stacks
Germany) was utilized for analysis. were performed with IMARIS imaging software.
Image analysis Image acquisition
All image analysis were processed and performed using We used a Leica SP5 confocal microscope for image
Leica Application Suite Advanced Fluorescence (Leica analysis. Excitations were 405 nm, 488 nm, 561 and 633 nm.
PLOS ONE | www.plosone.org 3 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
In addition we used an upright microscope with confocal crucial for blood brain barrier integrity in vivo [21] and in vitro
spinning disk system (QLC10 Visitech, Sunderland, UK) [12,17]. Cl-5 staining revealed a high number of Cl-5 positive
equipped with a temperature controlled chamber for live cell microvessels within COSC that were in close contact with
imaging (excitation: 488 nm, 568 nm). PDGFR beta positive cells (Figure 1 B). Another protein
expressed by pericytes is NG2. However NG2 is not solely
Induction of hypoxia and inflammation, DPI and Z- restricted to pericytes but also expressed e.g. by
DEVD-fmk treatment oligodendrocyte precursor cells [22]. Co-stainings of PDGFR
Hypoxia was induced by placing COSC membranes into beta and NG2 demonstrated a partial co-localization of both
medium which was bubbled with a gas mixture of 8% O , 5 % pericyte markers (arrowhead Figure 1 C) that was confirmed2
CO , rest N . Then probes were kept in a hypoxic incubator for by 3 D reconstructions (Movies S1, S2). Pericytes in native P32 2
24 hours. Atmosphere in the incubator was adjusted to reach mice (Figure 1 D) demonstrated a similar perivascular
levels of about 75 mmHg O  in culture medium. O  levels were localization as found in COSC (Figure 1 E, insets),2 2
monitored ahead and after hypoxia using an Oxylite pO demonstrating that our in vitro model recapitulates the2
sensing probe (Oxford Optronics, Oxford, UK). For anatomical in vivo situation. PDGFR beta positive cells
inflammation probes were incubated for 24 hours with medium appeared as cells with a DAPI positive cell soma, their
containing 10 or 100 ng/ml Interleukin 1 beta (IL1B R & D processes following microvessels (Figure 1 D, E insets). Note
Systems, Catalog number 401-ML). The inhibitor DPI was that the cell nucleus of a pericyte is mostly spared by PDGFR
preincubated for 1 hour at a concentration of 50 µmol/l ahead beta staining because PDGFR beta is a tyrosine kinase
of experimental manipulation. We chose this concentration situated in the membrane of a pericyte, not in the nucleus [22].
because we found DPI to be protective at 50 µmol/l on brain The pericyte nucleus is surrounded by PDGFR beta staining
endothelial cells during moderate hypoxic conditions that continues along pericyte processes.
(Zehendner et al. unpublished data). In addition DPI was Another antigen reported to be expressed by pericytes is
successfully used at 50 µmol/l by other groups[18]. To Desmin [22]. Desmin positive cells were found to co-localize
elucidate impact of caspase-3 inhibition on pericyte cell death with PDGFR beta next to cortical microvessels in COSC
cells were treated with Z-DEVD-fmk (20 µmol/l) during 24 h of (Figure 1 F). Desmin also co-localized with PDGFR beta in
moderate hypoxic and inflammatory (100 ng/ml IL1B) native P3 mice (Figure 1 G). In addition microvascular pericyte
conditions. Respective controls were treated with solvent (PBS, coverage, a hallmark for neurovascular integrity [4], was not
DMSO <0.1%) only. Hypoxic and / or inflammatory treatment significantly different at 5 DIV within COSC compared with
was started on DIV 2 and COSC were fixed for analyses on native P3 mice (COSC DIV5: 0.9 + 0.02 vs. P3: 0.89 + 0.02; n
DIV 3. = 93 - 101 microvessels [>10 COSC preparations, 6 native P3
mice], P = 0.7821, Figure 1 H).
Statistics
The neurovasular unit in the novel in vitro model
Results are documented as Mean + standard error of the
mean (SEM). All data were analyzed using Graphpad Prism for Immunohistochemical quadruple stainings show that in the
Windows (version 4.02, Graphpad, San Diego, CA, USA). presented in vitro approach almost all of the cell types of the
Datasets were checked for normalization by using D’Agostino neurovascular unit are present and display a preserved
and Pearson omnibus test. Groups that passed the D’Agostino morphology in COSC after 3 DIV. Using high power imaging of
and Pearson omnibus test[19] were therefore analyzed by two- cortical microvessels, pericytes and astrocytes we found that
sided unpaired student’s t-test. Groups that did not pass the the endothelium is enwrapped by pericytes that are further in
test for normalization were analyzed with a two sided Mann- close contact with astrocytes (Figure 2 A, Movie S3). NeuN
Whitney U test. Differences were regarded to be significant at immunoreactivity revealed the presence of neurons with an
alpha < 0.05. intact cellular morphology within the COSC preparation (Figure2 B). It has been reported that the basement membrane (BM),
which consists of laminins, proteoglykans, nidogens as well as
Results collagen IV [23], is of high relevance for neurovascular integrity
[24,25]. Therefore we were interested in the question if a BM is
Perivascular cells within cortical organotypic slice present in the cortical neurovascular in vitro model. We found
cultures that cortical microvessels and pericytes in COSC possess a
In the CNS the PDGFR beta receptor is exclusively BM as demonstrated by pan-Laminin stainings (antibody that
expressed by pericytes and therefore represents a sensitive binds to Laminin chains alpha-1, beta-1, alpha-2 and gamma-1,
pericyte marker [4,20]. To identify perivascular PDGFR beta and thereby many laminin isoforms that contain at least one of
positive cells in COSC we performed immunohistochemical these chain types) in COSC [24,26]. Laminins are
analyses with different vessel markers (CD105, Claudin 5), heterotrimers composed of an alpha, beta and gamma chain
PDGFR beta and DAPI. Our COSC preparations displayed a [23]. They are a hallmark for a proper formation of a BM and
well preserved cytoarchitecture of cortical layers after 5 days in therefore are a marker of the BM in microvessels in the CNS
vitro (DIV) as demonstrated by DAPI staining (Figure 1 A). The [25,27]. We observed a preservation of the BM in the
transmembraneous tight junctional protein Claudin 5 (Cl-5) is neurovascular unit of our in vitro preparation after 4 DIV
expressed by brain endothelial cells [17] in COSC and is (representative images from more than 10 COSC preparations,
PLOS ONE | www.plosone.org 4 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 1.  Pericytes in cortical organotypic slice cultures.  Microvessels and pericytes in COSC were immunolabelled with
different vascular markers (CD105, Cl-5), the pericyte marker PDGFR beta and DAPI. COSC preparations displayed a well
preserved cytoarchitecture of cortical layers after 5 DIV (DAPI, epifluorescent image, A). A high number of microvessels covered by
pericytes remained in COSC (epifluorescent image, B). PDGFR beta (green) partly co-localized (yellow) with the proteoglykan NG2
(red) in COSC which is another pericyte marker (arrowhead, confocal Z-stack with orthogonal section views, C). Similar to the
perivascular localization of PDGFR beta positive cells in newborn mice at the age of postnatal day 3 (D) we observed PDGFR beta
positive cells to be in close contact with microvessels in COSC after 5 DIV (E). Note the DAPI positive pericyte cell soma in which
the nucleus is mostly spared by PDGFR beta staining (marked by asterisks) and its processes that follow the microvessel (D, E
insets). Further the pericyte marker Desmin was found to be expressed by perivascular PDGFR beta positive cells in COSC (Panel
F) as well as in native P3 mice (Panel G). Pericyte coverage in cortical microvessels was not significantly different in native P3 mice
compared with COSC (H).
doi: 10.1371/journal.pone.0081637.g001
PLOS ONE | www.plosone.org 5 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 2.  The neurovascular unit in COSC.  Immunohistochemical stainings show a preserved morphology of astrocytes and
pericytes that cover cortical microvessels in COSC (A). Stainings with the neuronal marker NeuN demonstrated a preserved
neuronal morphology within the neurovascular unit of COSC (B).
doi: 10.1371/journal.pone.0081637.g002
Figure 3 A). High power confocal images revealed that cortical vesicular stomatitis glycoprotein-pseudotyped retroviruses to
microvessels and pericytes are embedded in the BM (Figure 3 transduce certain differentiation factors e.g. Mash1 and Sox2
B, Movie S4). Animated three-dimensional reconstructions with [29]. However, this process is crucially dependent on the
laser scanning confocal imaging confirmed these findings on capacity of pericytes to proliferate because a successful
an intact NVU (Movies S3, S4). transduction via retroviral vectors (except for lentiviruses) is
dependent on the ability of the targeted cells for cell division.
Live cell imaging of pericytes in EYFP-NG2 knockin We found that BrdU (exposition for 3 hours on DIV 3, 10 µmol/l)
mice was incorporated by some pericytes within 24 hours after
To address the question whether the in vitro model can be exposure (Figure 5 A, Movies S5, S6). In addition some
used for live cell imaging of pericytes, we labeled cortical pericytes were positive for Ki67 (Figure 5 B), a marker for cell
microvessels of P4 heterozygous EYFP-NG2 knockin mouse proliferation, after 4 DIV [30]. Ki67 and BrdU stainings were
mutants with tomato lectin. This mouse lineage carries an performed in more than 4 COSC preparations.
EYFP label inserted in exon 1 of the NG2 gene [28]. Here, in
line with our immunohistochemical NG2 analyses, the EYFP- Neuronal network activity and longterm survival of
NG2 signal was partly present next to cortical microvessels in pericytes in vitro
living COSC (Figure 4 A). However other NG2 cells were not To elucidate if COSC maintain spontaneous network activity
associated with microvessels indicating that they are not during culture conditions (5% CO2, 20% O2, rest N2, humidified
pericytes. Pericytes could be identified through the perivascular atmosphere, 37°C) multi-electrode array (MEA) recordings
EYFP-NG2 signal for up to 3 days after COSC preparation were performed as described previously with slight
(Figure 4 B). To verify that the obtained perivascular EYFP- modifications (Heck et al., 2008). Here we observed
NG2 signals were of pericyte origin we performed spontaneous synchronized neuronal network activity (Figure 6
immunohistochemical co-stainings of EYFP-NG2 cortices with A, representative image from n = 3 independent recordings,
PDGFR beta, NG2 and Cl-5, which demonstrated a co- DIV 5). Pericytes could be identified in COSC for up to 3
localization of EYFP-NG2 and PDGFR beta in a perivascular weeks. Figure 6 B shows a pericyte with characteristic
manner (Figure 4 C, D, insets). Co-stainings with NG2 and perivascular morphology after 14 days in vitro.
Cl-5 confirmed this observation (Figure 4 E).
Pericytes express elevated levels of caspase-3 after
Pericytes in COSC are capable of proliferation inflammation and hypoxia
It has recently been shown that pericytes from human origin Chronic hypoxia and inflammation have been proposed to be
can be reprogrammed into induced neuronal cells by the use of key factors leading to preterm brain injury [31]. In addition we
PLOS ONE | www.plosone.org 6 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 3.  Basement membrane in the neurovascular unit of COSC.  Co-labeling of cortical microvessels (Cl-5, green), pericytes
(PDGFR beta, red) and laminins with a pan-Laminin antibody (white) reveals the presence of a basement membrane (BM) in the
neurovascular unit in COSC after 4 DIV (confocal Z-stacks, maximum projection A). High power confocal magnification visualizes
the BM (arrowheads, B) that encloses microvessels (Cl-5, green) and pericytes (PDGFR beta, red). Note the DAPI positive cell
nuclei of the Cl-5 positive microvessel (asterisk) and the PDGFR beta positive pericyte (marked by #).
doi: 10.1371/journal.pone.0081637.g003
PLOS ONE | www.plosone.org 7 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 4.  Live cell imaging of pericytes in COSC from EYFP-NG2 mice.  Lectin stained microvessels in COSC appeared as red
labeled vascular structures that were surrounded by EYFP-NG2 expressing pericytes (A, insets). Pericytes could be identified for up
to 3 DIV 3 (B). Co-stainings with PDGFR beta (C, D), Cl-5 and NG2 demonstrate that perivascular EYFP-NG2 expressing cells are
indeed pericytes (E, insets).
doi: 10.1371/journal.pone.0081637.g004
and others have shown that caspase-3 is involved in neuronal in the developing brain [34], and moderate hypoxia result in
and BBB pathology during ischemia and inflammation caspase-3 activation in pericytes. Thus we subjected COSC
[17,32,33]. Therefore we were interested in the question if towards prolonged moderate hypoxia and different
inflammation induced by interleukin 1 beta (IL1B), a cytokine concentrations of IL1B. The cultures were exposed to a
which has been shown to disrupt proper white matter formation medium maintaining 71 + 2 mmHg pO2 for 24 hours. This pO2
PLOS ONE | www.plosone.org 8 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 5.  Pericytes in COSC are capable of cell division.  Confocal analyses revealed that BrdU (3 hours exposition, 10µmol/l)
was incorporated by pericytes within 24 hours on DIV 4 (arrowheads and asterisks in A mark a pericyte cell nucleus positive for
BrdU). Ki-67 is a marker for cell proliferation. Here, a pericyte cell nucleus immunoreactive for Ki-67 on DIV 4 is shown (arrowheads,
asterisks in B).
doi: 10.1371/journal.pone.0081637.g005
PLOS ONE | www.plosone.org 9 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 6.  Spontaneous neural network activity and longterm persistence of pericytes in COSC.  MEA recordings
demonstrated that spontaneous synchronized neural network activity is preserved in COSC under culture conditions on DIV 5 (A).
In addition pericytes were found to be present in COSC for weeks. Here, a 2 week old pericytes situated next to a Cl-5 positive
microvessel is shown (B).
doi: 10.1371/journal.pone.0081637.g006
 level is about 58% of control medium kept in normoxic cell It has been suggested that oxidative stress in form of
incubators (hypoxia 71 + 2.2 mmHg vs. normoxia 123 + 0.5 reactive oxygen species (ROS) plays a major role in immature
mmHg pO2, P < 0.0001, n = 15-48 measurements). brain damage [31]. Therefore we were interested in the
IL1B treatment resulted in significantly higher amounts of question if ROS may contribute to the observed enhancement
caspase-3 positive pericytes (representative image of a of caspase-3 cleavage in pericytes after 24 h of hypoxia and
caspase-3 positive pericyte treated for 24 hours with IL1B 100 whether an inhibition of oxidative stress may prevent
ng/ml, Figure 7 A) after 24 hours compared with control caspase-3 cleavage. Stainings for nitrotyrosine, a product that
(control: 1 + 0.1 vs. IL1B 10 ng/ml 1.42 + 0.15, P = 0.0269; arises from the reaction of the ROS peroxynitrite and the amino
control: 1 + 0.1 vs. IL1B 100 ng/ml 1.98 + 0.19, P < 0.0001, n = acid tyrosine [35] did not show any significant levels of
7 - 8 COSC preparations per group [535-668 pericytes], Figure nitrotyrosine in pericytes (representative image from at least 3
7 B). We detected significantly more caspase-3 positive COSC preparations, Figure 7 D, E). The NADPH oxidase is a
pericytes in COSC treated with IL1B 100 ng/ml than in those major source of reactive oxygen species [36] in cerebral
treated with 10 ng/ml (IL1B 10 ng/ml 1.42 + 0.15 vs. IL1B 100 ischemia and diphenyliodonium chloride (DPI) is a widely used
ng/ml 1.98 + 0.19, P = 0.0237, n = 8 COSC per group [535 - inhibitor of the NADPH oxidase [18]. We therefore tested if a
668 pericytes], Figure 7 B). pharmacological inhibition of NADPH oxidase by DPI has an
Moderate hypoxia for 24 hours resulted in a significantly
higher amount of cleaved caspase-3 positive pericytes (control: effect on blockade of caspase-3 cleavage in pericytes during
1 + 0.05 vs. hypoxia: 2.48 + 0.23, P < 0.0001, n = 11-41 COSC hypoxia. We found that DPI was not able to block caspase-3
preparations [571 - 2800 pericytes], Figure 7 C). Exposure of cleavage in pericytes. Despite DPI treatment caspase-3 levels
combined hypoxia and IL1B 100 ng/ml significantly elevated in the hypoxic group were significantly higher than in the
cleaved caspase-3 levels in pericytes (control: 1 + 0.05 vs. normoxic control (hypoxia+DPI: 3.049 + 0.16 vs. normoxic
yhpoxia+IL1B: 2.38 + 0.15, P < 0.0001, n = 18-41 COSC control: 1 + 0.07, P < 0.0001, n = 24 COSC preparations per
preparations [677-2800 pericytes], Figure 7 C). However, a group [1597-2812 pericytes], Figure 7 F).
combination of IL1B and 24 h of moderate hypoxia did not
increase caspase-3 cleavage compared with 24 h of moderate Caspase-3 inhibition ameliorates pericyte loss in
hypoxia alone (hypoxia: 2.48 + 0.23 vs. hypoxia+IL1B 2.38 + hypoxia and inflammation
0.15, P = 0.9763, n = 11 - 18 COSC preparations [571 - 677 Because we observed enhanced levels of cleaved caspase-3
pericytes], Figure 7 C). in pericytes upon ischemia and inflammation we were
PLOS ONE | www.plosone.org 10 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 7.  IL1B and moderate hypoxia induce caspase-3 cleavage in pericytes independently from peroxynitrite.  Exposition
of COSC towards pathologic conditions led to an increase of cleaved caspase-3 positive pericytes as shown in a representative
confocal image in panel A. Stimulation of COSC with different concentrations of IL1B resulted in significantly elevated levels of
cleaved caspase-3 in pericytes (B). Furthermore 24 hours of moderate hypoxia and a combination of IL1B and 24 hours of
moderate hypoxia resulted in caspase-3 cleavage in pericytes (C). Of note, combination of IL1B and hypoxia did not significantly
increase caspase-3 cleavage compared with probes treated with 24 hours of moderate hypoxia alone. Immunohistochemical
stainings for nitrotyrosine did not reveal any significant formation of this ROS reaction product in pericytes (D, E). The NADPH
oxidase inhibitor DPI [50µmol/l, 1 h pre-incubation] was not able to block caspase-3 cleavage under moderate hypoxic conditions.
*P<0.05, ***P<0.001.
doi: 10.1371/journal.pone.0081637.g007
interested in the question if caspase-3 activation may result in  0.05, P = 0.0142, n = 5-6 RFV from more than 3 COSC
pericyte loss. To address this question COSC were treated with preparations per group). COSC that were treated with Z-DEVD-
the selective and irreversible caspase-3 inhibitor Z-DEVD-fmk fmk sustained significantly higher pericyte numbers after
[17] during hypoxia and inflammation and pericyte numbers in hypoxic (hypoxia 0.31 + 0.02 vs. hypoxia/Z-DEVD-fmk 0.52 +
confocal z-stacks were compared with solvent treated normoxic 0.08, P = 0.0324, n = 5 RFV from more than 3 COSC) and
and non-inflammatory control COSC. We found that hypoxia inflammatory conditions compared with solvent treated groups
resulted in a significant decrease of pericytes/volume
compared with controls (control: 1 + 0.16 vs. hypoxia 0.31 + (IL1B 0.46 + 0.05 vs. IL1B/ZDEVD-fmk 0.95 + 0.16, P =
0.02, P = 0.0034, n = 5 - 6 RFV from more than 3 COSC). After 0.0406, n 5 - 8 RFV from more than 3 COSC preparations per
IL1B (100 ng/ml) treatment pericytes were also significantly group). However Z-DEVD-fmk was not able to completely block
reduced in cortical layers II-IV (control: 1 + 0.16 vs. IL1B 0.46 + pericyte loss under hypoxic conditions (control: 1 + 0.16 vs.
PLOS ONE | www.plosone.org 11 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
hypoxia/Z-DEVD-fmk 0.52 + 0.08, P = 0.0301, n = 5 - 6 RFV co-localized and varied in staining patterns. This is in good
from more than 3 COSC). agreement with previous reports [22]. The discrepancy e.g.
found in Desmin and PDGFR beta staining is due to the fact
Discussion that Desmin is a cytoskeleton protein which is situated
intracellular while PDGFR beta is a tyrosine kinase located on
In this report we demonstrate that cortical organotypic slice the pericyte’s cell surface (for a detailed review see for
cultures from neonatal rodents are a useful model to study example 22. In addition our electrophysiological recordings
pericytes within the intact neurovascular unit under various from COSC on MEA arrays demonstrate that spontaneous
experimental conditions. Further we show that pathological neuronal network activity is also preserved during incubator
conditions such as moderate hypoxia and inflammation result conditions and while recording in culture medium. We would
in caspase-3 mediated pericyte loss. like to stress this fact because usually MEA recordings in
To our knowledge no other experimental technique has yet COSC are performed using artificial cerebrospinal fluid
been described that allows the investigation of pericytes in vitro perfusion containing 95 % oxygen and 5 % CO2 [33].
for days including live cell imaging in a preserved Another goal of the present study was to establish a protocol
neurovascular environment. In line with our data Kovacs and for pericyte live cell imaging in the developing cortex in vitro
colleagues have shown that it is possible to study using a transgenic mouse line. In line with Karram et al. [28] we
neurovascular interactions in slice cultures of the hippocampus have shown that the EYFP-NG2 signal co-localizes partly with
of the neonatal rat [37] for days in vitro. Other in vitro the pericyte marker PDGFR beta in a perivascular manner and
approaches that imply pericytes are mostly used to assess verified that the EYFP signal is restricted to NG2 by the use of
BBB permeability during various conditions [38]. Such NG2 antibody staining. By lectin perfusion of the brain
approaches do not allow any investigations on microvasculature we demonstrate that pericytes can be
pathophysiological mechanisms in the immature cortex monitored live in vitro. Live cell imaging of pericytes within the
because an intact neural network is missing. However, the neurovascular unit may allow e.g. evaluation of the timing of
presented approach has certain limitations. Our in vitro setting pericyte constriction during hypoxia and/or inflammation. We
lacks cerebral blood flow. Shear stress in vessels has been would like to outline that for this purpose other genetically
shown to be important for a proper formation of the BBB [39]. modified mouse lines e.g. NG2DsRedBAC created by Zhu and
Therefore our model does not allow the analysis of shear colleagues [22,41] may also be useful.
stress effects on pericytes. This issue may be best addressed Hypoxia and inflammatory stimuli are hallmarks in the
by in vivo investigations because the pulsatile character of pathogenesis of immature brain damage [10]. The resulting
cerebral blood flow including the composition of the different encephalopathy of the premature is a severe brain injury with a
cellular blood components is difficult to reflect in vitro. Further heterogeneous phenotype [31]. Epidemiologic studies point to
live cell imaging with the presented lectin perfusion is not a relation of infections in prematurely born infants with
feasible for several weeks but rather limited to a couple of disorders like schizophrenia, autism [7,8,42] and later cognitive
days. The reason herefore is that the lectin signal vanishes impairment [43]. These pathologies are associated with cortical
over time because the dye is injected into circulation on the day gray matter changes [44,45]. In addition clinical studies have
of preparation. demonstrated a relation between premature brain injury and
Our data show that most of the cellular components of the cerebral blood flow (CBF) regulation: perturbances of CBF
NVU are morphologically preserved in vitro including a autoregulation is a predictor of severe brain injury in preterm
basement membrane which is an anatomical key feature for babies [46]. However, the exact cellular mechanisms
pericyte identification [40]. It has been suggested that e.g. underlying these phenomena are not fully understood. Peppiatt
during depletion of VEGF microvessels vanish, but the basal et al. [47] demonstrated that pericytes are capable of
laminina remains [40]. Our laminin stainings indicate that modulating the diameter of brain microvessels in slice
microvessels in cortical layers II-IV did not significantly preparations of the cerebellum. Further a loss of pericytes is
decrease within DIV 3-5 as nearly all BM tubes labeled by associated with the formation of cerebral microaneurysms [48].
laminin were associated with Cl5 positive microvessels. We These findings and recent data of Bell et al. who have shown
would like to stress that the presented model should be used that pericyte deficiency leads to neurovascular impairment and
within 3-5 DIV for quantitative pericyte analyses because we consecutive neuroinflammation in cortical layers [4], point to a
focused on this time period in our setting. We were not able to significant role of pericytes in regulating the NVU that is
distinguish an endothelial from a parenchymal BM with pan- important for cortical function throughout embryogenesis and
Laminin stainings, which is in line with data from Sixt et al. [25]. further development. Therefore another goal of the present
Endothelial and parenchymal BM only become distinguishable study was to evaluate the consequences of hypoxic and
from each other in conditions such as inflammation but not in inflammatory conditions on pericytes in the developing cortex.
physiological states. Therefore our model may also allow We demonstrate that pathological stimuli e.g. moderate
detailed analyses on the involvement of the BM during hypoxia over 24 hours and sterile inflammation by IL1B, result
pathological neurovascular conditions. We have used a set of in an increased level of cleaved caspase-3 activity within
pericyte, BM and vascular markers to be sure that PDGFR beta pericytes. We have previously shown that caspase-3
positive perivascular cells are indeed pericytes. Stainings for contributes to rapid anoxic neurovascular unit damage
the pericytes markers NG2, PDGFR beta and Desmin partly (RANUD) [17] and that in neonatal mice caspase-3 is essential
PLOS ONE | www.plosone.org 12 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
for proper cortical development [49], as already demonstrated matter impairment in the developing brain [9,34]. However our
before [50]. Furthermore it has been recently documented that data show that caspase-3 activation in the immature cortex
inhibition of caspase-3 reduces cortical lesions and improves under hypoxia and inflammation leads to pericyte loss. These
neurological outcome in a model of neonatal hypoxia-ischemia findings point out that pericytes may be another cell population
brain damage in rodents [51]. These data point to a crucial and which may be significantly affected by ischemic and
inflammatory insults in the developing cortex.
precisely regulated role of caspase-3 in physiology and Recently it has been shown that pericytes prevent
pathophysiology of the developing cortex. In line with other reperfusion of microvessels after MCAO in an oxidative stress
groups [51] our experiments show that caspase-3 cleavage is dependent manner. Here, the reactive oxygen species
present in hypoxic and inflammatory conditions in the immature peroxynitrite was shown to be of major importance [35]. In our
cortex. While it has been shown that a combination of model we did not find elevated levels of nitrotyrosine in
inflammation and hypoxia potentiates cortical damage [52], we pericytes after 24 hours of moderate hypoxia. Furthermore an
did not find an increase of caspase-3 cleavage in pericytes inhibitor of the NADPH oxidase (DPI) was also unable to block
upon combined inflammation and moderate hypoxia. However the observed caspase-3 activation in pericytes. Therefore we
our experimental setting differed from previous studies that conclude that peroxynitrite does not have a major role for
showed an increase of brain injury under these double hit caspase-3 activation in pericytes during 24 hours of hypoxia
without reoxygenation. However, in our setting we evaluated
conditions. For instance Brochu and colleagues [52] induced the effect of moderate hypoxia without reperfusion. This is a
hypoxia by exposing neonates to 8% O2 after ligation of one significantly different condition than the setting of Yemisci and
common carotid artery and injection with lipopolysaccharide colleagues who have investigated the effect of middle cerebral
(LPS), a surface protein of gram-negative bacteria that occur artery occlusion followed by reperfusion [35]. However,
for example during severe sepsis. These hypoxic and prolonged episodes of hypoxia in the newborn brain are
inflammatory stimuli are much more severe than moderate thought to better reflect the in vivo situation [9] because the
hypoxia and treatment with IL1B as used in the present study. majority of prematurely born infants do not suffer from focal
Another explanation could be that the neurodegenerative ischemia, e.g. an occluded cerebral vessel, but rather from
mechanism of LPS may be mediated by direct activation of the insufficient blood oxygenation due to immature lungs and an
brain endothelium and independently from IL1B as shown by immature breathing center within the brain stem [55].
Our results on the capacity of pericytes to proliferate in vitro
Murray et al. [53]. However it has recently been proposed that may represent a promising therapeutical strategy. It has been
moderate inflammation by IL1B and moderate hypoxic shown that pericytes are involved in central nervous scar
conditions are clinically more relevant for the pathogenesis of formation [56]. In addition pericytes may be a target to
brain injury of prematurely born infants [54]. Our results compensate for neuronal injury [29]. Karow and colleagues
indicate that caspase-3 cleavage in pericytes increases in a have shown that human pericytes can be reprogrammed into
dose dependent manner. Even rather low concentrations of induced neuronal cells [29]. These authors used a retroviral
IL1B (10 ng/ml) were capable of inducing significant caspase-3 vector to transduce differentiation factors which requires that
activation in pericytes which points to a critical vulnerability of targeted cells are capable of cell division [29]. BrdU and Ki-67
pericytes in the developing cortical neurovascular unit. Our stainings show that some pericytes in our experimental
results indicate that hypoxia and inflammation lead to approach remain their capacity to proliferate. Hence we
speculate that the presented technique may lead to protocols
caspase-3 activation which results in pericyte loss. A for the reprogramming of pericytes into neuronal cells in a
pharmacological inhibition with a specific caspase-3 inhibitor preserved neurovascular unit within the cortex. We hypothesize
support this hypothesis. Pericyte loss has been shown to result that such protocols may also be of use for regenerative
in BBB impairment and causes an accumulation of neurotoxic therapies in order to compensate for neuronal injury in the
substances in the cortical parenchyma in vivo [4]. We suspect cortex e.g. stroke or dementia. In addition we found that
that the detected pericyte loss in our experimental conditions pericytes remain in COSC for up to three weeks which gives
may have major implications for pathophysiological processes the opportunity to perform e.g. long term fate mapping of
in vivo within the developing cortex (Figure 8) and may result in pericytes. However this will of course require different
e.g. accumulation of neurotoxins that may cause cognitive transgenic mouse models than we used in our experiments
decline in ageing [4]. We suggest that caspase-3 inhibition may (see for example 56.
be a promising target with regard to stabilization of the NVU by
protecting pericytes in inflammatory and hypoxic states. This
finding may also have implications for other parts of the CNS
e.g. the spinal cord. Hypoxic brain damage and inflammation
have been demonstrated to be predominantly relevant in white
PLOS ONE | www.plosone.org 13 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Figure 8.  Pericyte impairment in the developing cortex during pathologic conditions.  Hypoxia and inflammation result in
caspase-3 dependent loss of pericytes in cortical layers. Pericyte loss may be of relevance for perturbances in neurovascular
integrity in the developing brain and may be involved in a variety of pathologic sequelae e.g. cognitive decline in ageing or
neuroinflammation.
doi: 10.1371/journal.pone.0081637.g008
PLOS ONE | www.plosone.org 14 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Supporting Information are enclosed by laminin (pan-laminin labeling, white) which
indicates the presence of a basement membrane.
Movie S1.  Here, an animated confocal 3 D Z-stack (AVI)
(thickness of planes 700nm) of a quadruple staining of
pericytes, cell nuclei and microvessels is shown. White: Movie S5.  Here a cell nucleus (DAPI, blue) positive for
NG2 a marker for pericytes, Red: vascular marker CD105, BrdU (green) of a pericyte (labeled with PDGFR beta in red)
Green: PDGFR beta, a pericyte marker, DAPI in blue: stains is shown in an animated confocal 3 D Z-stack (700nm
cellular nuclei. Note that NG2 and PDGFR beta partly co- thickness of planes). Note that the cell membrane of the
localize in a perivascular manner indicating that PDGFR beta pericyte stains positive for PDGFR beta (red) whilst its cell
positive perivascular cells are indeed pericytes. nucleus is spared due to the localization of PDGFR beta on the
(AVI) cell membrane of the pericyte.
(AVI)
Movie S2.  To better appreciate the NG2 (white) signal
within the perivascular pericyte channels representing the Movie S6.  To better demonstrate the pericyte’s BrdU
microvessel (red) and PDGFR beta signal (green) from positive cell nucleus from Movie S5 all channels except
Movie S1 have been removed in this reconstruction. Note the green channel (BrdU) have been removed in this 3 D
the cell nucleus of the NG2 positive pericyte in the center of the reconstruction.
animation that is in close contact with the NG2 signal. (AVI)
(AVI)
Acknowledgements
Movie S3.  A 3 dimensional reconstruction (confocal Z-
stack, thickness of planes: 700nm) of the neurovascular The authors want to thank Sabine Rickheim-Lowack and Beate
unit. Red: PDGFR beta positive pericyte which covers a Krumm for their excellent technical assistance. Jerome Mordel
cortical microvessel (labeled with Cl-5, green). Note the close was of great help for the MEA recordings. We are thankful for
relationship of astrocytotic endfeet (GFAP, white) that surround the excellent support by the IMB Core Facility Microscopy of
the pericyte. An animated fly-through the vessel demonstrates the Institute of Molecular Biology (IMB) in Mainz for confocal
that PDGFR beta is partly co-localizing (yellow) with Cl-5 which image acquisition and processing. Heterozygous EYFP-NG2
is due to the fact the PDGFR beta is a tyrosine kinase located mice were kindly provided by the lab of Prof. Trotter (University
at the cell membrane while Cl-5 is a transmembraneous tight of Mainz) and we are most thankful for the provided help.
junction linker protein.
(AVI) Author Contributions
Conceived and designed the experiments: CMZ. Performed the
Movie S4.  An animated 3 D reconstruction (confocal Z- experiments: CMZ HEW. Analyzed the data: CMZ HEW.
stack, thickness of planes: 700nm) of the neurovascular Contributed reagents/materials/analysis tools: CMZ HJL. Wrote
unit. Red: PDGFR beta positive pericyte next to a cortical the manuscript: CMZ HJL.
microvessel (green, Cl-5). Note that both vessel and pericyte
References
1. Attwell D, Buchan AM, Charpak S, Lauritzen M, Macvicar BA et al. 8. Rapoport JL, Giedd JN, Gogtay N (2012) Neurodevelopmental model
(2010) Glial and neuronal control of brain blood flow. Nature 468: 232– of schizophrenia: update 2012. Mol Psychiatry 17: 1228–1238. doi:
243. doi:10.1038/nature09613. PubMed: 21068832. 10.1038/mp.2012.23. PubMed: 22488257.
2. Zlokovic BV (2008) The Blood-Brain Barrier in Health and Chronic 9. Hagberg H, Peebles D, Mallard C (2002) Models of white matter injury:
Neurodegenerative Disorders. Neuron 57: 178–201. doi:10.1016/ comparison of infectious, hypoxic-ischemic, and excitotoxic insults.
j.neuron.2008.01.003. PubMed: 18215617. Ment Retard Dev Disabil Res Rev 8: 30–38. doi:10.1002/mrdd.10007.
3. Winkler EA, Bell RD, Zlokovic BV (2011) Central nervous system PubMed: 11921384.
pericytes in health and disease. Nat Neurosci 14: 1398–1405. doi: 10. Volpe JJ, Kinney HC, Jensen FE, Rosenberg PA (2011) The
10.1038/nn.2946. PubMed: 22030551. developing oligodendrocyte: key cellular target in brain injury in the
4. Bell RD, Winkler EA, Sagare AP, Singh I, LaRue B et al. (2010) premature infant. Int J Dev Neurosci 29: 423–440. doi:10.1016/
Pericytes control key neurovascular functions and neuronal phenotype
in the adult brain and during brain aging. Neuron 68: 409–427. doi: j.ijdevneu.2011.02.012. PubMed: 21382469.
10.1016/j.neuron.2010.09.043. PubMed: 21040844. 11. Thal SC, Luh C, Schaible E-V, Timaru-Kast R, Hedrich J et al. (2012)
5. Ballabh P, Braun A, Nedergaard M (2004) The blood-brain barrier: an Volatile Anesthetics Influence Blood-Brain Barrier Integrity by
overview: structure, regulation, and clinical implications. Neurobiol Dis Modulation of Tight Junction Protein Expression in Traumatic Brain.
16: 1–13. doi:10.1016/j.nbd.2003.12.016. PubMed: 15207256. Injury - PLOS ONE 7: e50752. doi:10.1371/journal.pone.0050752.
6. Indredavik MS, Vik T, Evensen KAI, Skranes J, Taraldsen G et al. 12. Zehendner CM, Luhmann HJ, Kuhlmann CRW (2009) Studying the
(2010) Perinatal risk and psychiatric outcome in adolescents born neurovascular unit: an improved blood-brain barrier model. J Cereb
preterm with very low birth weight or term small for gestational age. J Blood Flow Metab 29: 1879–1884. doi:10.1038/jcbfm.2009.103.
Dev Behav Pediatr 31: 286–294. doi:10.1097/DBP. PubMed: 19638997.
0b013e3181d7b1d3. PubMed: 20431402. 13. Stoppini L, Buchs PA, Muller D (1991) A simple method for organotypic
7. Meyer U, Feldon J, Dammann O (2011) Schizophrenia and Autism: cultures of nervous tissue. J Neurosci Methods 37: 173–182. doi:
Both Shared and Disorder-Specific Pathogenesis Via Perinatal 10.1016/0165-0270(91)90128-M. PubMed: 1715499.
Inflammation? Pediatr Res 69: 26R–33R. doi:10.1203/PDR. 14. Nimmervoll B, Denter DG, Sava I, Kilb W, Luhmann HJ (2011) Glycine
0b013e318212c196. PubMed: 21289540. receptors influence radial migration in the embryonic mouse neocortex.
PLOS ONE | www.plosone.org 15 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
Neuroreport 22: 509–513. doi:10.1097/WNR.0b013e328348aafe. 34. Favrais G, van de Looij Y, Fleiss B, Ramanantsoa N, Bonnin P et al.
PubMed: 21666516. (2011) Systemic inflammation disrupts the developmental program of
15. Karram K, Goebbels S, Schwab M, Jennissen K, Seifert G et al. (2008) white matter. Ann Neurol 70: 550–565. doi:10.1002/ana.22489.
NG2-expressing cells in the nervous system revealed by the NG2- PubMed: 21796662.
EYFP-knockin mouse. Genesis 46: 743–757. doi:10.1002/dvg.20440. 35. Yemisci M, Gursoy-Ozdemir Y, Vural A, Can A, Topalkara K et al.
PubMed: 18924152. (2009) Pericyte contraction induced by oxidative-nitrative stress impairs
16. Guedel E (1937) INHALATION ANESTHESIA: A Fundamental Guide. capillary reflow despite successful opening of an occluded cerebral
American Journal of the Medical Sciences. 120 p. Available: http:// artery. Nat Med 15: 1031–1037. doi:10.1038/nm.2022. PubMed:
journals.lww.com/amjmedsci/Citation/1937/07000/ 19718040.
INHALATION_ANESTHESIA__A_Fundamental_Guide.23.aspx. 36. Kahles T, Luedike P, Endres M, Galla H-J, Steinmetz H et al. (2007)
Accessed 25 February 2013 NADPH oxidase plays a central role in blood-brain barrier damage in
17. Zehendner CM, Librizzi L, de Curtis M, Kuhlmann CRW, Luhmann HJ experimental stroke. Stroke 38: 3000–3006. doi:10.1161/STROKEAHA.
(2011) Caspase-3 contributes to ZO-1 and Cl-5 tight-junction disruption 107.489765. PubMed: 17916764.
in rapid anoxic neurovascular unit damage. PLOS ONE 6: e16760. doi: 37. Kovács R, Papageorgiou I, Heinemann U (2011) Slice Cultures as a
10.1371/journal.pone.0016760. PubMed: 21364989. Model to Study Neurovascular Coupling and Blood Brain Barrier In
18. Hansel CM, Zeiner CA, Santelli CM, Webb SM (2012) Mn(II) oxidation Vitro. Cardiovascular Psychiatry and Neurology 2011: 1–9 doi:
by an ascomycete fungus is linked to superoxide production during 10.1155/2011/646958.
asexual reproduction. Proc Natl Acad Sci U S A 109: 12621–12625. 38. Nakagawa S, Deli MA, Kawaguchi H, Shimizudani T, Shimono T et al.
doi:10.1073/pnas.1203885109. PubMed: 22802654. (2009) A new blood-brain barrier model using primary rat brain
19. D’Agostino RB, Pearson ES (1990). Testing for Departures from endothelial cells, pericytes and astrocytes. Neurochem Int 54: 253–263.
Normality. 44: 316–321. doi:10.1016/j.neuint.2008.12.002. PubMed: 19111869.
20. Winkler EA, Bell RD, Zlokovic BV (2010) Pericyte-specific expression of 39. Cucullo L, Hossain M, Puvenna V, Marchi N, Janigro D (2011) The role
PDGF beta receptor in mouse models with normal and deficient PDGF of shear stress in Blood-Brain Barrier endothelial physiology. BMC
beta receptor signaling. Mol Neurodegener 5: 32. doi: Neurosci 12: 40. doi:10.1186/1471-2202-12-40. PubMed: 21569296.
10.1186/1750-1326-5-32. PubMed: 20738866. 40. Krueger M, Bechmann I (2010) CNS pericytes: concepts,
21. Nitta T, Hata M, Gotoh S, Seo Y, Sasaki H et al. (2003) Size-selective misconceptions, and a way out. Glia 58: 1–10. doi:10.1002/glia.20898.
loosening of the blood-brain barrier in claudin-5-deficient mice. J Cell PubMed: 19533601.
Biol 161: 653–660. doi:10.1083/jcb.200302070. PubMed: 12743111. 41. Zhu X, Bergles DE, Nishiyama A (2008) NG2 cells generate both
22. Armulik A, Genové G, Betsholtz C (2011) Pericytes: Developmental, oligodendrocytes and gray matter astrocytes. Development 135: 145–
Physiological, and Pathological Perspectives, Problems, and Promises. 157. doi:10.1242/dev.004895. PubMed: 18045844.
Dev Cell 21: 193–215. doi:10.1016/j.devcel.2011.07.001. PubMed: 42. Brown AS (2012) Epidemiologic studies of exposure to prenatal
21839917. infection and risk of schizophrenia and autism. Dev Neurobiol, 72:
23. Hallmann R, Horn N, Selg M, Wendler O, Pausch F et al. (2005) 1272–6. doi:10.1002/dneu.22024. PubMed: 22488761.
Expression and function of laminins in the embryonic and mature
vasculature. Physiol Rev 85: 979–1000. doi:10.1152/physrev. 43. Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJS (2002)
00014.2004. PubMed: 15987800. Cognitive and behavioral outcomes of school-aged children who were
24. Enzmann G, Mysiorek C, Gorina R, Cheng Y-J, Ghavampour S et al. born preterm: a meta-analysis. JAMA 288: 728–737. doi:10.1001/jama.
(2013) The neurovascular unit as a selective barrier to 288.6.728. PubMed: 12169077.
polymorphonuclear granulocyte (PMN) infiltration into the brain after 44. Olabi B, Ellison-Wright I, McIntosh AM, Wood SJ, Bullmore E et al.
ischemic injury. Acta Neuropathol 125: 395–412. doi:10.1007/ (2011) Are there progressive brain changes in schizophrenia? A meta-
s00401-012-1076-3. PubMed: 23269317. analysis of structural magnetic resonance imaging studies. Biol
25. Sixt M, Engelhardt B, Pausch F, Hallmann R, Wendler O et al. (2001) Psychiatry 70: 88–96. doi:10.1016/j.biopsych.2011.01.032. PubMed:
Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles 21457946.
in T cell recruitment across the blood-brain barrier in experimental 45. Wen W, He Y, Sachdev P (2011) Structural brain networks and
autoimmune encephalomyelitis. J Cell Biol 153: 933–946. doi:10.1083/ neuropsychiatric disorders. Curr Opin Psychiatry 24: 219–225. doi:
jcb.153.5.933. PubMed: 11381080. 10.1097/YCO.0b013e32834591f8. PubMed: 21430538.
26. Agrawal S, Anderson P, Durbeej M, van Rooijen N, Ivars F et al. (2006) 46. Osborn DA, Evans N, Kluckow M (2003) Hemodynamic and antecedent
Dystroglycan is selectively cleaved at the parenchymal basement risk factors of early and late periventricular/intraventricular hemorrhage
membrane at sites of leukocyte extravasation in experimental in premature infants. Pediatrics 112: 33–39. doi:10.1542/peds.
autoimmune encephalomyelitis. J Exp Med 203: 1007–1019. doi: 112.1.33. PubMed: 12837865.
10.1084/jem.20051342. PubMed: 16585265. 47. Peppiatt CM, Howarth C, Mobbs P, Attwell D (2006) Bidirectional
27. Hagg T, Portera-Cailliau C, Jucker M, Engvall E (1997) Laminins of the control of CNS capillary diameter by pericytes. Nature 443: 700–704.
adult mammalian CNS; laminin-alpha2 (merosin M-) chain doi:10.1038/nature05193. PubMed: 17036005.
immunoreactivity is associated with neuronal processes. Brain Res 48. Lindahl P (1997) Pericyte Loss and Microaneurysm Formation in
764: 17–27. doi:10.1016/S0006-8993(97)00419-8. PubMed: 9295189. PDGF-B-Deficient Mice. Science 277: 242–245. doi:10.1126/science.
28. Karram K, Goebbels S, Schwab M, Jennissen K, Seifert G et al. (2008) 277.5323.242. PubMed: 9211853.
NG2-expressing cells in the nervous system revealed by the NG2- 49. Heck N, Golbs A, Riedemann T, Sun J-J, Lessmann V et al. (2008)
EYFP-knockin mouse. Genesis 46: 743–757. doi:10.1002/dvg.20440. Activity-dependent regulation of neuronal apoptosis in neonatal mouse
PubMed: 18924152. cerebral cortex. Cereb Cortex 18: 1335–1349. doi:10.1093/cercor/
29. Karow M, Sánchez R, Schichor C, Masserdotti G, Ortega F et al. bhm165. PubMed: 17965127.
(2012) Reprogramming of Pericyte-Derived Cells of the Adult Human 50. Kuan CY, Roth KA, Flavell RA, Rakic P (2000) Mechanisms of
Brain into Induced Neuronal Cells. Cell Stem Cell 11: 471–476. doi: programmed cell death in the developing brain. Trends Neurosci 23:
10.1016/j.stem.2012.07.007. PubMed: 23040476. 291–297. doi:10.1016/S0166-2236(00)01581-2. PubMed: 10856938.
30. Schonk DM, Kuijpers HJ, van Drunen E, van Dalen CH, Geurts van 51. Chauvier D, Renolleau S, Holifanjaniaina S, Ankri S, Bezault M et al.
Kessel AH et al. (1989) Assignment of the gene(s) involved in the (2011) Targeting neonatal ischemic brain injury with a pentapeptide-
expression of the proliferation-related Ki-67 antigen to human based irreversible caspase inhibitor. Cell Death Dis 2: e203. doi:
chromosome 10. Hum Genet 83: 297–299. doi:10.1007/BF00285178. 10.1038/cddis.2011.87. PubMed: 21881605.
PubMed: 2571566. 52. Brochu M-E, Girard S, Lavoie K, Sébire G (2011) Developmental
31. Ferriero DM (2004) Neonatal Brain. Injury - New England Journal of regulation of the neuroinflammatory responses to LPS and/or hypoxia-
Medicine 351: 1985–1995. doi:10.1056/NEJMra041996. ischemia between preterm and term neonates: An experimental study.
32. Karatas H, Aktas Y, Gursoy-Ozdemir Y, Bodur E, Yemisci M et al. J Neuroinflammation 8: 55. doi:10.1186/1742-2094-8-55. PubMed:
(2009) A nanomedicine transports a peptide caspase-3 inhibitor across 21599903.
the blood-brain barrier and provides neuroprotection. J Neurosci 29: 53. Murray CL, Skelly DT, Cunningham C (2011) Exacerbation of CNS
13761–13769. doi:10.1523/JNEUROSCI.4246-09.2009. PubMed: inflammation and neurodegeneration by systemic LPS treatment is
19889988. independent of circulating IL-1β and IL-6. J Neuroinflammation 8: 50.
33. Nimmervoll B, White R, Yang J-W, An S, Henn C et al. (2013) LPS- doi:10.1186/1742-2094-8-50. PubMed: 21586125.
Induced Microglial Secretion of TNF Increases Activity-Dependent 54. Volpe JJ (2011) Systemic inflammation, oligodendroglial maturation,
Neuronal Apoptosis in the Neonatal. Cerebral Cortex - Cerebral Cortex and the encephalopathy of prematurity. Ann Neurol 70: 525–529. doi:
23: 1742–1755. doi:10.1093/cercor/bhs156. 10.1002/ana.22533. PubMed: 22028217.
PLOS ONE | www.plosone.org 16 November 2013 | Volume 8 | Issue 11 | e81637
Pericytes in the Developing Cortex In Vitro
55. Darnall RA, Ariagno RL, Kinney HC Darnall RA, Ariagno RL, Kinney 10.1016/j.clp.2006.10.004. doi:10.1016/j.clp.2006.10.004. PubMed:
HC (2006) The late preterm infant and the control of breathing, sleep, 17148011.
and brainstem development: a review. Clin Perinatol 33: 883–914; 56. Göritz C, Dias DO, Tomilin N, Barbacid M, Shupliakov O et al. (2011) A
abstract. doi:10.1016/j.clp.2006.10.004. PubMed: 17148011 x doi: Pericyte Origin of Spinal Cord Scar Tissue. Science 333: 238–242. doi:
10.1126/science.1203165. PubMed: 21737741.
PLOS ONE | www.plosone.org 17 November 2013 | Volume 8 | Issue 11 | e81637