THE PUBLISHING HOUSE MEDICIN E OF THE ROMANIAN ACADEMY Research article PERIADVENTITIAL CELLS OF MENINGEAL AND INTRAPARENCHYMAL ARTERIES AND… [625362]

THE PUBLISHING HOUSE MEDICIN E
OF THE ROMANIAN ACADEMY Research article
PERIADVENTITIAL CELLS OF MENINGEAL AND INTRAPARENCHYMAL
ARTERIES AND ARTERIOLES, CHARACTERIZED
BY IMMUNOHISTOCHEMISTRY AND ULTRASTRUCTURAL EXAMINATION
OF HUMAN FRESH AND POSTMORTEM TISSUE
Leon DANAILA1 and Dorel Eugen A RSENE2
1Neurosurgery Department, Na tional Institute of Neurology and Neurovascular Diseases, Bucharest, Romania ,
2Patholog y Department, National Institute of Neurology and Neurovascular Diseases, Bucharest, Romania
Corresponding author : Dorel Eugen Arsene, MD, PhD, Pathologic Anatomy Depart ment,
National Institute of Neurology and Neurovascular Diseases
E-mails: [anonimizat] ; [anonimizat]
Accepted
The small arterial branches of the brain surface and penetrating the nervous parenchyma display some
peculiar peripheral cells coating their surface. These cells are not entirely identified regarding their
structure or function. We intended to study them thoroughly using a large pane l of
immunohistochemical antibodies and transmission electron microscopy on both postmortem and fresh
human tissue s. In the arteries located within the subarachnoid space, t he periadventitial cells showed
strongest positivity for vimentin, α-smooth muscle actin, collagen type IV, PDGFR -β and PDGFR -α.
Lower positivity was found for TE -7 fibroblast marker, S100 protein, CD68, CD63, CD133, and
caveolin -1. For the arteries located within the brain parenchyma, the strongest reaction retained was
only for vimenti n and collagen type IV. In both locations, these periadventitial cells were negative for
NOTCH -1, CD13, DDR1, FLK1, C D45, nestin, laminin, desmin, or DDR2. Our results suggest that
these cells are myofibroblasts with limited stem cell characteristics and m ost probably involved in
angiogenesis . Furthermore , we found that t he arteries within the subarachnoid space were not
enveloped by pia mater, unlike those pen etrating the brain parenchyma .
Key words : human brain , arteries , adventitia , myofibroblasts , pericytes.
INTRODUCTION1
The c ells envelop ing the arteries within the
central nervous system are rarely characterized in
depth. The se cells , which we named periadventitial
cells (PAC s), are located at the outer limit of the
vessel, either positioned on a thin collagen layer, or
embedded in this one (Fig. 1). PAC s could have
several origins, functions and structure. We
intended to unravel the fundamental structural
details of PAC s at molecular level . The peripheral
portion of cerebral vessels seems to be involv ed in
many processes, the perivascular drainage of
deleterious compounds ( i.e. beta-amyloid) being an
important one1,2,3,4. This is in close relationship with
the lymphatic c learance of the brain5,6, subsequently
stated as being involved in some neurodegen erative
conditions7,8. Since they are located i n this
peripheral position, PAC s, apparently originating in

1Proc. Rom. Acad., Series B, 2018 , 20(1) , p. 23–32 the pia mater, are described as being a structural
part of the pial -glial basement membrane4. These
PAC s are, however, insufficiently characterized i n
humans and until now immunohistochemistry
played only a modest role in this characterization
process . Therefore, a more detailed approach to
their structure and functions seemed appropriate.
MATERIAL S AND METHODS
Since PAC s could be simple fibroblasts, w e tried
to verify this hypothesis first. TE7 is an antibody
especially designed to react with human fibroblasts,
while not cross reacting with other connective tissue
cells9. Also, the antibody seems to be more specific
than other ones conceived for the sa me pur pose,
1B10 and 5B510. To test the inclusion of these cells
in the fibroblast category, DDR1, DDR2, PDGFRα,
PDGFRβ and collagen IV staining were performed.
DDR1 ( discoidin domain receptor 1 ) is a receptor

24 Leon D ănăila and Dorel Eugen Arsene
interacting with all types of collagen (from t ype I to
type V)11. It is involved in tissue homeostasis, cell
proliferation, differentiation, adhesion, m igration
and invasion12. DDR2 ( discoidin domain receptor
2) is also involved in cancer d evelopment and
metastasi s. DDR1 is expressed by epithelial cel ls,
whereas DDR2 appears in mesenchymal cells14,15.
DDR2, as well as PDGFRα, are known as
fibroblasts cell membrane components, at lea st in
the cardiac ones16. Both PDGFRα and PDGFRβ
have increased expressions on fibroblasts
membrane, related to their proliferation potential
and transform ation into myofib roblasts17. Collagen
IV, in conjunction with laminin, is a fundamental
component of tissue specific basement membranes,
including the vascular ones18. It is also known as
being involved in myofibroblasts developm ent and
migration19.
On the other hand, PAC s could also be
considered as having progenitor cell potential , as
already described20. To this purpose, we performed
immunostainings for CD34, c -kit (CD117), Flk -1
(VEGFR -2) and CD133 , molecules known as being
expressed by progenitor cells . Some studies a lso
define globally the meninge s as a stem cell niche21,
so we tested this hypothesis on PAC s using CD133,
CD45 and NOTCH -1 staining22, in conjunction with
the rest of the panel of antibodies already mentioned
(Flk-1, CD34, or c -kit). Nestin is also mentio ned as
being present in the meninges, which were
considered as being a source of stem cells23,
therefore we also stained the tissue for this marker.
For the potential inclusion of PAC s in the
pericyte category, we used markers currently
described as being expre ssed by this cell type24, as
α-smooth muscle actin, nestin, vimentin, and
platelet derived growth factor receptor -beta
(PDGFR -β), respectively. All these are currently
used to identify pericytes25. Also, desmin and
alanyl -aminopeptidase (CD13) are bo th mentioned
as useful pericyte markers26.
Since PAC s appear as exceedingly elongated
cells, their possible relationship with the so -called
telocytes27-35 could not be ignored. To test this
hypothesis, positive staining for c -kit protein
(CD117), CD34, PDG FR-α, CD63, smooth muscle
actin and vimentin would have been considered as
significant, even though some of them are also
expressed by pericytes. In this case, the cell shape
and localization as well as its ultrastructural details
were taken into account.
CD34 is a transmembrane protein expressed
mostly on hematopoietic progenitor cells but also on several primitive pluri potential stem cells36.
Since it labels the majority of gastrointestinal
strom al tumors (GIST)37 which seem to originate in
telocytes, we also used it as a marker in our series.
CD117, also known as c -kit, is a receptor tyrosin
kinase protein encoded by KIT gene , expressed on
multipotent progenitor cells and hematopoietic stem
cells. It is mentioned as being useful for identifyin g
vascular s tem cells22, while also identifying reliably
the mast cells38. Both CD34 and CD117 are also
considered as ma rkers for telocytes29. Flk -1
(VEGFR -2) is a receptor tyrosin kinase expressed
on progenitor cells, with potential role in
vasculogenesis and neuroge nesis39. CD133 , on the
other hand, is currently used for identification of
stem cells in both normal and tumor tissue40. CD45,
also known as common leuk ocyte antigen (CLA), is
a membrane glyco protein expressed mostly on
leukocytes, as its name suggests, bu t also in
circulating cancer stem cells41 and in the v essel wall
stem cells22. Nestin, an intermediate filament, is
expressed in various cell types, including mostly
multipotent neural stem cells42 but also in
endothelial stem cells, even though it seems n ot to
be present in mature vasculature43. Notch -1 is a
transmembrane receptor mentioned as marker for
vascular stem cells22, being involved in stem cell
processes as survival, differentiat ion or
proliferation44.
CD63 is a tetraspanin protein related to
exosomes45. Caveolin -1, also known to b e present in
telocytes46, was tested in our panel of antibodies.
Since some studies included these vascular
coating cells (probably PAC) in the category of pial
cells, without supplem entary specifications47, we
also deci ded to compare the immunoprofiles of the
two categories (PAC and pial cells) in order to see if
identity is found.
PATIENTS
We analyzed the structure of cerebral vessel s in
a series of 20 patients , 13 men (65%) and 7 women
(35%) , ranging from 44 to 85 year s old (mean
65.35) deceased in the neurology and neurosurgery
departments of our institute. Informed consent was
obtained i n each case, from the patients or their
relatives. The work was approved by the Ethic
Committee of the institute according to interna tional
requirements. Complete autopsies were performed
in each deceased patient . The interval between
death and tissue sampling varied from 6 hours to 36
hours. The major inclusion criteri on was the

Periadventitial cells of brain arteries and arterioles 25
macroscopic integrity of the brain surface and
leptomenin ges at macroscopic examination. Cases
were excluded when presenting cerebral
pathological conditions which could significantly
change the structure of the brain and leptomeninges,
as infarcts or hemorrhages with superficial
involvement , tumors or neuro degenerative diseases.
After a few weeks fixation in 10% buffered
formalin , the brain was sectioned in each subject in
1 cm wide slices . Previously, in each case we
harvested the whole leptomeningeal coating one
hemisphere and embedded i t in paraffin as a sin gle
block. We worked on archival paraffin tissue blocks
prepared in this manner and retrieved from the
archive of our institute. Both the arachnoid blocks
as well as those with brain fragments taken
randomly from cortical areas of the frontal, parietal,
and occipital lobes were sectioned and stained with
Haematoxylin and Eosin for microscopic
examination . All available clinical and pathological
data (resulting from autopsy specimen examination)
were retrospectively examined. Separately, in two
patients unde rgoing surgical procedures for
meningioma ablation (female s, age d 50 and
65 respectively ), small fragments of vessels and
adjacent nervous parenchyma were sampled for
electron microscopy.
HISTOPATHOLOGY
AND IMMUNOHISTOCHEMISTRY
All samples processed for histopathology were
analyzed for the expression of several
immunoh istochemical markers . Two micrometer –
thick sections were obtained from the previously
histologically examined paraffin block s. Each
section was deparaffiniz ed and hydrated in graded
ethanol concentrations. Heat induced antigen
retrieval was obtained with the buffer indicated by
each antibody supplier for 30’. The slides were then
treated with 3% hydrogen peroxide for 20’ at room
temperature to block endogenous peroxidase
activity. Tissue slid es were incubated with primary
antibodies solution overnight at room temperature,
followed by two PBS (phosphate buffered saline)
changes. The list of primary antibodies used is
shown below (see the table) . Epitope Source Dilution

Vimentin Dako, Glost rup, Denmark 1:50
Desmin Dako, Glostrup, Denmark 1:50
TE7 Merck, Millipore, Billerica, CA,
USA 1:50
CD13 Leica Biosystems, Newcastle
upon Tyne, UK 1/100
PDGFRα NeoMarkers, Fremont, CA,
USA 1:40
Laminin Sigma -Aldrich, St Louis, MI,
USA 1/1000
Nestin Santa Cruz Biotechnology, CA,
USA 1:50
DDR1 Santa Cruz Biotechnology, CA,
USA 1:50
DDR2 Santa Cruz Biotechnology, CA,
USA 1:50
αSMA Dako, Glostrup, Denmark 1/100
S100
protein Dako, Glostrup, Denmark 1/500
CD63 Leica Biosystems, Newcastle
upon Tyne, UK 1:50
CD117 Dako, Glostrup, Denmark 1/400
Caveolin Santa Cruz Biotechnology, CA,
USA 1:50
CD45 Leica Biosystems, Newcastle
upon Tyne, UK 1/100
NOTCH -1 Santa Cruz Biotechnology, CA,
USA 1/100
CD133 MyBioSource, San Diego, CA,
USA 1/500
Collagen
type IV Leica Biosystems, Newcastle
upon Tyne, UK 1/100
The reaction was visualized using Poly -HRP –
GAM/R/R IgG detection kit for 30’ (Immunologic,
Duiven, The Netherlands) using 3,3’ -diamino –
benzidine, followed by counterstaining with
hematoxylin. Negative cont rol was obtained by
omitting the primary antibodies and replacing them
with PBS (phosphate buffered saline) . Positive
control was also available for each antibody
according to manufacturer’s instructions.
Images were taken using a system with an
Olympus B5 1 light microscope with attached
Olympus SP -350 camera, while being further
acquired , measured and annotated using the
QuickPhoto MICRO 2.2 software ( PROMICRA,
Prague, Czech Republic). We examined all
category of vessels – arteries, arterioles, veins,
venules, capillaries – in the subarachnoid space, on
their trajectory entering the brain and within the
nervous parenchyma. Focus was set on arterial and
arteriolar vessels , since veins, venules and

26 Leon D ănăila and Dorel Eugen Arsene
capillaries possess a vascular wall too thin to enable
a cor rect identification of a distinct adventitia or its
structural components . Cells located at the vascular
periphery , at the outer limit of adventitia, were ou r
target, regarding their shape, distribution, and
immuno phenotype (Fig. 1). We describe d these
cells as periadventitial cells (PAC s).

Figure 1. In (A), a large artery (440 µm in diameter) within the
subarachnoid space is cut longitudinally. Several elongated
cells, whose nuclei are highlighted by arrows, are coating the
vascular surface. (B) and (C). High power view of two
intraparenchymal arteries, at cortical -subcortical limit (B) and
within the white matter (C) show the same elongated cells at
their surface; arrows point their nuclei. (Hematoxylin and Eosin
– (A). Original magnification 200 ×; Mallo ry trichrome – (B),
(C). Original magnification 1000 ×).
We graded the immuno staining extension and
intensity as following: 0 for no reaction; 1 for
isolated cells around the vessel circumference
showing strong positivity or alternatively, a weak
positivity on most of the s ame section area; and 2
for strong , diffuse staining of all or almost all these
cells. Since all antibodies were supposed to be
expressed in a cytoplasmic or membran ous manner
(no nuclear ones were used ), these were the
expressions we cons idered in our examination.
Separate examination was performed for
subarachnoid arteries /arterioles , intraparenchymal
arteries/arterioles and pia cells.
ELECTRON MICROSCOPY
In the two cases of patients undergoing surgery,
the fresh tissue composed of small fragments of
nervous paren chyma and small vessels adjacent to
the lesion were harvested and immediately
immersed in glutaraldehyde. Transmission electron
microscopy was performed on small (1 mm3) tissue fragments processed according to a routine Epon
embed ding procedure. Thin sections (50 –60 nm)
were examined with a Morgagni 286 transmission
electron microscope (FEI Company, Eindhoven,
The Netherlands) at 80 kV.
RESULTS
The histological findings were uniform across
the whole series, in all regions of brain and within
the entire leptomeningeal tissue , regardless of the
patient’s age or subjacent pathology .
The periadventitial cells (PAC s) within the
subarachnoid space were strongly positive for
vimentin and collagen IV , α-smooth muscle actin,
PDGFRα , and PDGFRβ . Vimentin showed strong
positivity in the cytoplasm of almost all PAC s
coating arteries of all calibers within the
subarachnoid space, as well as the muscular layer
(Figs. 2 and 3).

Figure 2. Vimentin is expre ssed almost in a continuous manner
in PACs of a subarachnoid large artery. PACs are located in
peripheral position or embedded in the adventitial collagen
layer (arrows). A perforating artery also have a separate coating
layer with positive staining for vi mentin (short arrows).
(Vimentin immunostaining, 200 × original magnification).

Figure 3. Details of an intraparenchymal artery sectioned
longitudinally. PACs are obvious at the external border of the
vessel (arrows). (Vimentin immunostaining, 400 × origin al
magnification).

Periadventitial cells of brain arteries and arterioles 27

Figure 4. Collagen type IV is also expressed in all PACs around
a large artery from the subarachnoid space, in a single or double
layer at this point of section (Collagen IV immunostaining,
200× original magnification).

Figure 5. Col lagen IV is also expressed at intraparenchymal
level, but seems to represent a pial coating. ( A) – pia mater
(arrows) is following a penetrating artery. (B) – two sections of
arterial profiles show the same coating positive for collagen IV
(arrow). (Collag en IV immunostaining, 400 × original
magnification – (A); 200 × original magnification – (B).

Figure 6. Smooth muscle actin is expressed in an almost
continuous layer in PACs of a subarachnoid large artery ( A). In
(B), several arterial profiles within the brain parenchyma show
actin expressed mostly with a granular pattern, either coating
the vessel wall or slightly separated from it (arrows). (α -smooth
muscle actin immunostaining. 200 × – (A), and 400 × – (B),
original magnifications).

Figure 7. ( A). PDGFRα is strongly exp ressed in an almost
continuous layer in PACs coating a 300 µm artery of the
subarachnoid space, mostly embedded in a thick collagen
matrix (long arrows) but also at the outmost periphery (short
arrow). ( B) shows almost the same aspect (long arrows).
Howeve r, a smaller arterial branch is negative, as is the pia
adjacent to it (short arrow). ( C) shows an intraparenchymal
arterial branch with only scattered PACs showing positivity.
(PDGFRα immunostaining, 400 × original magnification).

Figure 8. ( A). CD133 is expressed in several PACs, coating
larger arteries in the subarachnoid space (long arro ws).
However, many peripheral areas are devoid of positivity (short
arrow, up left). In ( B), CD133 shows a more granular positivity
in PACs of an intraparenchymal arterial branch (CD133
immunostaining, 400 × original magnification).

Figure 9. ( A). PDGFRβ expression is seen almost in the same
strong manner as that of PDGFRα (long arrows). Here,
PDGFRβ is also expressed in PACs of the smaller artery (short
arrow). Pia mater is still negative. ( B).
Two small arterial intraparenchymal profiles express PDGFRβ
in a zonal manner (PDGFRβ immunostaining,
400× original magnification) .

28 Leon D ănăila and Dorel Eugen Arsene

Figure 10. TE7 is expressed in an almost continuous pattern in
PACs of a perforating artery (arrows). Inset: detail of positive
cells coating the vessel. (TE7 immunostaining, 200 × original
magnification, inset – 1000 × original magnification) .

Figure 11. (A). The expression of CD68 is restricted to limited
areas of PACs on a subarachnoid artery (arrows). ( B). Within
the parenchyma, several PACs express CD68, with a granular
pattern, mostly adjacent to the nucleus (CD68 immunostaining,
200× original magnification – (A),
400× original magnification – (B).

Figure 12. The overall reactivity of PACs in the subarachnoid
space. Maximal staining was obtained for vimentin, collagen
type IV , α-smooth muscle actin, PDGFRα, and PDGFRβ.
(Abbreviation: sub=subarachnoid) .

Figure 13. The reactivity of PACs in the cerebral parenchyma.
Unlike the superficial vessels, the intraprenchymal ones were
strongly positive only for vimentin and collagen ty pe IV .

Figure 14. Ultrastructural details of PACs. In ( A), the elongated
cell is visibly embedded in a dense collagen matrix (arrow).
In (B), the image is composed of two sections, showing the
nuclear and perinuclear details of PACs, with numerous
mitoc hondria and sparse endoplasmic reticulum.
No obvious junctions are visible with the surrounding
extracellular structures (Electron microscopy) .
Collagen type IV was also strongly positive in
the cytoplasm of PAC s at this level, the cellular
processes givi ng an almost continuous pattern,
organized on one or two layers, outside and within
the collagen coating of the arterial wall (Fig 4). The
same profile was obtained with α -smooth muscle
actin (Fig . 6 A), PDGFR -α (Fig . 7 A, B), and
PDGFR -β (Fig . 9 A).
A w eaker staining was seen for CD133, TE7,
CD45 (CLA), CD63, caveolin and S100 protein. As
an example, CD133 positivity stained only scattered
PAC s coating large arteries, while large areas were
devoid of positivity (Fig . 8 A). CD68 was expressed
only by rare PAC s, coating the arteries within the
subarachnoid space, in a strong manner, but limited
to the cell body , adjacent to the nucleus and
apparently sparing the cellular processes (Fig . 11 A).
For PAC s located inside the brain parenchyma,
the strongest reac tion was obvious for collagen IV
and vimentin, whereas PDGFRβ, TE7 , PDGFRα,
α-smooth muscle actin, CD45, CD63, CD133 and
S100 protein were only mildly expressed. A

Periadventitial cells of brain arteries and arterioles 29
representative picture of vimentin expression of
PAC s within the parenchyma is showed (Fig . 3).
For the collagen type IV expression within the
parenchyma, representat ive images are s hown in
Figure 5. However, it wa s obvious that pi a mater is
strongly positive for this marker and that the coating
layer around perforating arteries at cortical lev el is
composed of a n extension of this pia material
(Fig. 5 A). Arterial profiles in more profound zones
of the brain keep the same structure, with a coating
layer at certain distance from the arterial wall,
probably determined by a certain degree of cereb ral
edema, present in almost all cases, due to their
subsequent pathological states (Fig . 5 B). For
α-smooth muscle actin, the reaction of PAC s was
different at this level, with an obvious coating layer
apposed on the arterial wall and local separation in
other areas (Fig . 6 B). Almost the same aspect was
visible for CD133, with the important difference
that the staining was more granular, dispersed
within the cytoplasm of PAC s (Fig. 8 B). PDGFRα
(Fig. 7 C) and PDGFRβ (Fig . 9 B) were expressed
in rare, sca ttered PAC s around small arteries within
the parenchyma. The fibroblast marker TE7 was
present as a continuous layer around the
intraparenchymal arteries, practically surrounding
the vessels (Fig . 10). CD68 expression was more
consistent at this level, wit h numerous PAC s
showing strong positivity, although in a
discon tinuous arrangement (Fig . 11 B). No reaction
was present for CD13, CD34, CD117, FLK 1,
NOTCH -1, DDR1, DDR2, nestin, laminin or
desmin, in either of the two vascular locations,
while maintaining varying positivity in other cells,
as endothelium, smooth muscle cells, scattered
neurons or astrocytes (internal control) . As a
consequence, we ignored all these positive cells and
considered only our target ed structure s. Caveolin -1
showed a non -specific staining on all arterial
structures in the subarachnoid space, and no
positivity within the parenchymal arteries. S100
protein , in exchange, was positive in all arterial
structures and pia mat er, thus we decided not to take
into account both reactions (not shown). The overall
immunophenotype in our series is summarized in
Figure s 12 and 13 .
The ultrastructure of PAC s showed elongated
shape s, with extremely long processes, cytoplasmic
organelles but lacking conspicuous membrane
junctions with the extracellu lar matrix (Fig. 14). DISCUSSION
We tried to evaluate the peripheral adventitial
cells (PAC s) surrounding the brain vessels from an
immunohistochemical and ultrastructural point of
view. Subarachnoid PAC s were only weakly
positive for CD133, S100 protein, and the fibroblast
marker TE7. They showed strong positivity for
vimentin, collagen type IV, α-smooth muscle actin ,
PDGFR β, and PDGFRα. PAC s around smaller
arteries within the parenchyma were weakly
positive for CD133, S100 protein, smooth muscle
actin, PD GFRβ, PDGFRα , and TE7, and strongly
only for vimentin, and collagen type IV. On the
other hand, pia cells were weakly positive for CD34
and S100 protein, while showing strong staining
with antibodies against vimentin and collagen
type IV.
No positive rea ction of PAC s was seen for
nestin , CD13 or desmin . These antibodies are
considered as being characteristic markers for
pericytes24-26. However, the ir expression within the
central nervous system is not strictly limited to a
certain cell, being dynamically regulated as a
consequence of various injuries and also a shift
between cell populations at this level is possible26.
As an example, fibroblast markers can be expre ssed
by pericytes48. Many cell types, including
fibroblasts and smooth muscle cells, express both
PDGFRα and PDGFRβ . PAC s expressed in both
locations these two molecules, therefore a certain
mix of fibroblast and smooth muscle cell
immunoprofile (myofibroblast) is probable for
them , furthermore reinforced by the strong reaction
for α -smooth muscl e actin.
Globally, despite their elongated phenotype , the
meningeal cells of the pia mater and those ones
surrounding the brain vessels – PAC s – were
phenotypically various in our study. To summarize ,
they expressed almost equally markers for
fibroblasts ( TE7, collagen IV), pericytes ( α-smooth
muscle actin, PDGFR β), telocytes (PDGFR α,
α-smooth muscle actin) and less for progenitor cells
(CD133). Vimentin was also strongly expressed, but
without specificity, since it is considered as a
common element for all these cell categories. The
strong reaction of PACs to PDGFRα in our study
could only suggest a certain relation of these cells to
telocytes, which are know n to express this
marker49,50. On the other hand, CD34 and CD117,
considered as the essential markers for
characterizing telocyt es51,52 were negative in our
study.

30 Leon D ănăila and Dorel Eugen Arsene
The strong expression of PDGFRα suggests a
certain involvement of these cells in angiogenesis
by recruiting other perivascular cells and
proliferation if needed, th rough the action of
PDGF53. We found α-smooth muscle ac tin to be
positive in multiple PAC s, whereas pia mater was
completely negative for this molecule . This
suggests that PAC s are not a n extension of the pia
layer , even though former structural studies (based
only on ultrastructural examination ) considered th e
two cell populations as a continuum54.
Also, the positivity for α-smooth muscle actin is
described as being present in telocytes, a common
feature shared by pericytes51. Telocytes are
described at vascular level only in medium sized
arteries, and limited to the endothelial surface of the
vessel , not as adventitial cell s55. At cerebral level,
they were detected in the meninges within dura
mater (at least in d ogs), close to capillaries35. Since
telocytes are still difficult to be characterized in
terms of immunophenotype56, the cells we identified
as PAC s could be conceiv ably described as
telocyte s (according to their ultrastructural
features) , but exhibiting a particular phenotype ,
supposedly demanded by local structural and
functional conditions . PACs could also be involved ,
at this level , in modulati on of the local home ostasis,
as a rudimental second nervous network32.
The putative scavenger role of perivascular cells
was affirmed in older studies and on animal
models57. To verify this, we performed CD68
staining, with inconclusive results. This wa s
because the vast majo rity of meningeal and
parenchymal PAC s were devoid of CD68 reactivity ,
except for some very rare oval shaped cells,
suggesting possibly perivascular macrophages .
Some studies found that pericytes are able to
transform and migrate as microglia withi n the
nervous parenchyma58 and that they are actively
involved in the immune processes of the brain59. In
this regard, the presence of CD68 positive cells in
our series, in both the vascular periphery (only as
macrophages) and within the adjacent parenchyma
(as ac tivated microglia) has not confirm ed this
potential association. The role of some perivascular
cells in local repairing mechanisms as well as their
presumed potential of motility60 remain , however,
under debate until much larger series of patients
will be examined with more complex methods .
On the other hand, the structure and function of
pia mater itself are still not conclusive. Early studies
using cell cultures speculated the potential identity
of pia and arachnoid cells, since not normal but tumor ( meningioma ) cells were used61.
Ultrastructural studies on human meninges rarely
supplemented by rudim ental immunostaining62
affirm that the coating of cerebral vessels is
composed of pia mater54,63,64. The perivascular
space between pia mater and the brain vessels is
considered as being critical in the proc ess of brain
metastasis65. This perivascular compartment is also
thought to be a way for lymphati c drainage of the
brain66. PAC s could play a regulatory role in this
unusual transport system of brain fluids. This could
occur in conjunction with the extremely complex
elements composing the vascular matrisome
(extracellular matrix and associated proteins)67.
PDGF ac tivity is conditioned by the presence of
its receptors PDGFR -α and PDGFR -β, respectively,
inducing angiogenesis and modulating the
proliferation and recruiting of perivascular cells53.
Since PAC s in the subarachnoid space expressed
both marker s in our stud y, a certain potential for
angiogen etic activity could be presumed , demanded
by pathological situations as ischemic or tumoral
conditions.
Our study has some shortcomings, such as a
relative ly narrow age category of subjects. Extreme
age category as very y oung ( i.e. infant cases) or
near centenary ones would have been necessary to
compare the overall phenotypic expression of
PACs . However, no significant differences were
visible among the cases included in our series,
regardless their age. Another potential bias
generator is the variability in the tissue processing
steps. Such parameters, as postmortem delay,
different fixation time s, and fixative concentration,
paraffin embedding materials and temperature,
which are not us ually standardized68 could affect t he
immunohistochemical reactivity of specific cell
types from a ca se to another69. However, in our
series there was a remarkable uniform immune –
phenotype among the same cell type within the
same location (arterial coating ), regardless of all
these potenti ally confounding factors .
CONCLUSIONS
Peripheral cells of the adventitial layer possess a
complex phenotype, suggesting their potential
inclusion in several cellular categories: fibroblasts,
pericytes and possibl y telocyte s. Overall, they are
most probabl y myofibroblasts with limited stem cell
potential and strong angiogen ic capabilities. A
remarkable uniformity of these findings was

Periadventitial cells of brain arteries and arterioles 31
obvious regardless the age , pathological state
among subjects or region of the brain where the
vessels were located . Their scavenger role was not
confirmed . With certainty , the arteries and arterioles
within the brain parenchyma seemed to be
surrounded by a n extension of pia mater, while
those one s within the subarachnoid space were not,
even though the shape of their coating c ells was
almost similar within the two locations .
REFERENCES
1. Hawkes C.A., Jayakody N., Johnston D.A., Bechmann I.,
and Carare R.O. Failure of perivascular drainage of β –
amyloid in cerebral amyloid angiopathy. Brain Pathol.
2014 ; 24, 396 -403.
2. Morris A.W., Carare R.O., Schreiber S., and Hawkes C.A.
The Cerebrovascular Basement Membrane: Role in the
Clearance of β -amyloid and Cerebral Amyloid Angiopathy.
Front Aging Neurosci. 2014 ; 6, 251.
3. Keable A., Fenna K., Yuen H.M., Johnston D.A., Smyth
N.R., Smith C., et al. Deposition of amyloid β in the walls
of human leptomeni ngeal arteries in relation to perivascular
drainage pathways in cerebral amyloid angiopathy. Biochim
Biophys Acta 2016 ; 1862, 1037 -1046.
4. Morris A.W., Sharp M.M., Albargothy N.J., Fernandes R.,
Hawkes C.A., Verma A., et al. Vascular basement
membranes as p athways for the passage of fluid into and
out of the brain. Acta Neuropathol. 2016 ; 131, 725 -736.
5. Bakker E.N., Bacskai B.J., Arbel -Ornath M., Aldea R.,
Bedussi B., Morris A.W., et al. Lymphatic Clearance of the
Brain: Perivascular, Paravascular and Signif icance for
Neurodegenerative Diseases. Cell Mol Neurobiol. 2016 ; 36,
181-94.
6. Engelhardt B., Carare R.O., Bechmann I., Fluge l A., Laman
J.D., and Weller R.O. Vascular, glial, and lymphatic
immune gateways of the central nervous system. Acta
Neuropathol. 2016; 132, 317 -338.
7. Carare R.O., and Kalaria R. Cerebrovascular pathology: the
dark side of neurodegeneration. Acta Neuropathol. 2016 ;
131, 641 -643
8. Tarasoff -Conway, J.M., Carare, R.O., Osorio, R.S., Glodzik
L., Butler T., Fieremans E., et al. Clearance syst ems in the
brain –implications for Alzheimer disease. Nat Rev Neurol.
2016 ; 11, 457-470.
9. Pilling D., Fan T., Huang D., Kaul B., and Gomer RH.
Identification of Markers that Distinguish Monocyte –
Derived Fibrocytes from Monocytes, Macrophages, and
Fibroblas ts. PLoS ONE 2009 ;4, e7475.
10. Goodpaster T., Legesse -Miller A., Hameed M.R., Aisner
S.C., Randolph -Habecker J., and Coller H.A. An
Immunohistochemical Method for Identifying Fibroblasts
in Formalin -fixed, Paraffin -embedded Tissue. J Histochem
Cytochem. 2008 ; 56, 347 –358.
11. Vogel W., Gish G.D., Alves F., and Pawson T. The
discoidin domain receptor tyrosine kinases are activated by
collagen. Mol Cell . 1997 ; 1:13–23
12. Moreau V., and Saltel F. (Type I collagen fibrils and
discoidin domain receptor 1 set invadosomes straight. Mol
Cell Oncol. 2015 ; 2, e1004963. 13. Fan Y., Xu Z., Fan J., Huang L., Ye M., Shi K., et al.
Prognostic significance of discoidin domain receptor 2
(DDR2) expression in ovarian cancer. Am J Transl Res.
2016 ; 8, 2845 -2850.
14. Leitinger B. Transmembran e collagen receptors. Annu Rev
Cell Dev Biol . 2011 ; 27, 265 –290.
15. Leitinger B. Discoidin domain receptor functions in
physiological and pathological conditions. Int Rev Cell Mol
Biol. 2014 ; 310, 39–87
16. Ivey M.J., and Tallquist M.D. Defining the Cardiac
Fibro blast. Circ J . 2016 ; 80, 2269 -2276.
17. Fan B., Ma L., Li Q., Wang L., Zhou J., and Wu J. Role of
PDGFs/PDGFRs signaling pathway in myocardial fibrosis
of DOCA/salt hypertensive rats. Int J Clin Exp Pathol .
2013 ; 7, 16-27. eCollection 2014.
18. Mueller J., Gaertner F.C., Blechert B., Janssen K.P., Essler
M. Targeting of tumor blood vessels: a phage -displayed
tumor -homing peptide specifically binds to matrix
metalloproteinase -2-processed collagen IV and blocks
angiogenesis in vivo. Mol Cancer Res. 2009 ; 7, 1078 -1085.
19. Loscertales M., Nicolaou F., Jeanne M., Longoni M., Gould
D.B., Sun, Y., et al. Type IV collagen drives alveolar
epithelial -endothelial association and the morphogenetic
movements of septation. BMC Biol. 2016 ; 14, 59.
20. Torsney E., Hu Y ., Xu, Q. Adventitial progenitor cells
contribute to arteriosclerosis. Trends Cardiovasc Med.
2005 ; 15, 64-68.
21. Decimo I., Fumagalli G., Berton V., Krampera M., Bifari,
F. Meninges: from protective membrane to stem cell niche.
Am J Stem Cells 2012 ; 1, 92-105.
22. Orlandi A. The Contribution of Resident Vascular Stem
Cells to Arterial Pathology. Int J Stem Cells. 2015 ; 8, 9-17
23. Decimo I., Bifari F., Rodriguez F.J., Malpeli G., Dolci S.,
Lavarini V., et al. Nestin – and doublecortin -positive cells
reside in adult sp inal cord meninges and participate in
injury -induced parenchymal reaction. Stem Cells 2011 ; 29,
2062 -2076.
24. Fisher M. Pericyte signaling in the neurovascular unit.
Stroke 2009 ; 40[Suppl 1], S13-S15.
25. Bergers G., Song S. The role of pericytes in blood -vesss el
formation and maintenance. Neuro -Oncology 2005 ; 7, 452-
464
26. Fernández -Klett F., Priller J. The fibrotic scar in
neurological disorders. Brain Pathol. 2014 ; 24, 404 -413.
doi: 10.1111/bpa.12162.
27. Popescu L.M., Gherghiceanu M., Cretoiu D., Radu E. The
connec tive connection: interstitial cells of Cajal (ICC) and
ICC-like cells establish synapses with immunoreactive
cells. Electron microscope study in situ. J Cell Mol Med.
2005 ; 9, 714 –730
28. Popescu L.M., Faussone -Pellegrini M.S. Telocytes – a case
of serendipity : the winding way from interstitial cells of
Cajal (ICC), via interstitial Cajal -like cells (ICLC) to
telocytes. J Cell Mol Med. 2010 ; 14, 729 –740
29. Ardeleanu C., Bussolati G. Telocytes are the common cell
of origin of both PEComas and GISTs: an evidence –
supported hypothesis. J Cell Mol Med. 2011 ; 15, 2569 –
2574.
30. Popescu B.O., Gherghiceanu M., Kostin S., Ceafalan L.,
Popescu L.M. Telocytes in meninges and choroid plexus.
Neurosci Lett. 2012 ; 516, 265 -269
31. Luesma M.J., Gherghiceanu M., Popescu L.M. Telocytes
and stem cells in limbus and uvea of mouse eye. J Cell Mol
Med. 2013 ; 17, 1016 -1024.

32 Leon D ănăila and Dorel Eugen Arsene
32. Smythies J., Edelstein L. Telocytes, exosomes, gap
junctions and the cytoskeleton: the makings of a primitive
nervous system? Front Cell Neurosci. 2014 ; 7, 278.
33. Bosco C., Díaz E., Gutiérrez R., Gonzalez J., Parra -Cordero
M., Rodrigo R., et al. A putative role for telocytes in
placental barrier impairment during preeclampsia. Med
Hypotheses 2015 ; 84, 72-77.
34. Bei Y., Wang F., Yang C., Xiao J. Telocytes in regenerative
medicin e. J Cell Mol Med. 2015 ; 19, 1441 -1454. (a) .
35. Xu T., Lu S., Zhang H. Transmission electron microscope
evidence of telocytes in canine dura mater. J Cell Mol Med.
2016 ;20, 188 -192.
36. Ordonez N.G. Immunohistochemical Endothelial Markers:
A Review. Adv Anat Pat hol. 2012 ; 19, 281 –295
37. Pusztaszeri M.P., Seelentag W., Bosman, F.T.
Immunohistochemical expression of endothelial markers
CD31, CD34, von Willebrand factor, and Fli -1 in normal
human tissues. J Histochem Cytochem. 2006 ; 54, 385 -395.
38. Bujan B., Mennel H.D. E levated number of mast cells in
angiomatous meningioma. Clin Neuropathol. 2015 ; 34, 54-
56
39. Yang X., Cepko C.L. Flk -1, a receptor for vascular
endothelial growth factor (VEGF), is expressed by retinal
progenitor cells. J Neurosci. 1996 ; 16, 6089 -6099.
40. Wu Y., Wu P.Y. CD133 as a marker for cancer stem cells:
progresses and concerns. Stem Cells Dev. 2009 ; 18, 1127 –
1134.
41. Kantara C., O'Connell M.R., Luthra G., Gajjar A., Sarkar
S., Ullrich R.L., et al. Methods for detecting circulating
cancer stem cells (CCSCs) a s a novel approach for
diagnosis of colon cancer relapse/metastasis. Lab Invest.
2015 ; 95, 100 -112.
42. Park D., Xiang A.P., Mao F.F., Zhang L., Di C.G., Liu
X.M., et al. Nestin is required for the proper self -renewal of
neural stem cells. Stem Cells. 2010 ;28, 2162 -2171.
43. Suzuki S., Namiki J., Shibata S., Mastuzaki Y., Okano, H.
The neural stem/progenitor cell marker nestin is expressed
in proliferative endothelial cells, but not in mature
vasculature. J Histochem Cytochem. 2010 ;58, 721 -730.
44. Matsui W.H. Cance r stem cell signaling pathways.
Medicine (Baltimore) 2016 ; 95(1 Suppl 1) , S8-S19.
45. Pols M.S., Klumperman J. Trafficking and function of
tetraspanin CD63. Exp Cell Res. 2009 ; 315,1584 –1592 .
46. Zhang H.Q., Lu S.S., Xu T., Feng Y.L., Li H., Ge J.B.
Morphological evidence of telocytes in mice aorta. Chin
Med J (Engl). 2015 ; 128, 348 -352.
47. Pollock H., Hutchings M., Weller R.O., Zhang E.T.
Perivascular spaces in the basal ganglia of the human brain:
their relationship to lacunes. J Anat. 1997 ; 191, 337 -346
48. Rustenhove n J., Scotter E.L., Jansson D., Kho D.T.,
Oldfield R.L., Bergin P.S., et al. An anti -inflammatory role
for C/EBPδ in human brain pericytes. Sci Rep . 2015 ;5,
12132.
49. Zhou Q., Wei L., Zhong C., Fu S., Bei Y., Huica R.I., et al.
Cardiac telocytes are double p ositive for CD34/PDGFR -α.
J Cell Mol Med. 2015 ; 19, 2036 –2042
50. Xiao J., Chen P., Qu Y., Yu P., Yao J., Wang H., et al.
Telocytes in exercise -induced cardiac growth. J Cell Mol
Med. 2016 ; 20, 973 -979.
51. Bei Y., Zhou Q., Fu S., Lv D., Chen P., Chen Y., et al.
Cardiac telocytes and fibroblasts in primary culture:
different morphogenesis and immunophenotypes. PLoS
One 2015 ; 10, e0115991 (b) . 52. Arafat E.A. Ultrastructural and immunohistochemical
characteristics of telocytes in the skin and skeletal muscle
of newborn rats. Acta Histochem. 2016 ; 118, 574 -580.
53. Raica M., Cimpean A.M. Platelet -Derived Growth Factor
(PDGF)/PDGF Receptors (PDGFR) Axis as Target for
Antitumor and Antiangiogenic Therapy. Pharmaceuticals
2010 ;3, 572 -599.
54. Zhang E.T., Inman C.B., Weller R.O. In terrelationships of
the pia mater and the perivascular (Virchow -Robin) spaces
in the human cerebrum. J Anat. 1990 ; 170,111-123.
55. Li H., Lu S., Liu H., Ge J., Zhang H. Scanning electron
microscope evidence of telocytes in vasculature. J Cell Mol
Med. 2014 ; 18, 1486 –1489 .
56. Cretoiu D., Radu B.M., Banciu A., Banciu D.D., Cretoiu
S.M. Telocytes heterogeneity: From cellular morphology to
functional evidence. Semin Cell Dev Biol . 2016 ;64, 26-39
57. Kida S., Steart P.V., Zhang E.T., Weller R.O. Perivascular
cells act as scavengers in the cerebral perivascular spaces
and remain distinct from pericytes, microglia and
macrophages. Acta Neuropathol. 1993 ;85, 646 -652.
58. Özen I., Deierborg T., Miharada K., Padel T., Englund E.,
Genove G., et al. Brain pericytes acquire a microgl ial
phenotype after stroke. Acta Neuropathol. 2014 ; 128,
381-96.
59. Hurtado -Alvarado G., Cabañas -Morales A.M., Gómez –
Gónzalez B. Pericytes: brain -immune interface modulators.
Front Integr Neurosci. 2014 ; 7, 80.
60. Edelstein L., Smythies J. Epigenetic aspects of
telocytes/cordocytes: jacks of all trades, masters of most.
Front.Cell.Neurosci. 2014 ; 8:32.
61. Feurer D.J., Weller R.O. Barrier functions of the
leptomeninges: a study of normal meninges and
meningiomas in tissue culture. Neuropathol Appl
Neurobiol. 1991 ; 17, 391 -405.
62. Alcolado R., Weller R.O., Parrish E.P., Garrod D. The
cranial arachnoid and pia mater in man: anatomical and
ultrastructural observations. Neuropathol Appl Neurobiol.
1988 ; 14, 1-17.
63. Hutchings M., Weller R.O. Anatomical relationships of the
pia mater to cerebral blood vessels in man. J Neurosurg.
1986 ; 65, 316 -325.
64. Nicholas D.S., Weller R.O. The fine anatomy of the human
spinal meninges. A light and scanning electron microscopy
study. J Neurosurg. 1988 ; 69, 276 -282.
65. Saito N., Hatori T., Murata N., Zhang Z.A., Nonaka H.,
Aoki K., et al. Comparison of metastatic brain tumour
models using three different methods: the morphological
role of the pia mater. Int J Exp Pathol. 2008 ; 89, 38-44.
66. Diem A.K., Tan M., Bressloff N.W., Hawkes C., Morris
A.W., W eller R.O., et al. A Simulation Model of
Periarterial Clearance of Amyloid -β from the Brain. Front
Aging Neurosci. 2016 ; 8:18.
67. Joutel A., Haddad I., Ratelade J., Nelson M.T.
Perturbations of the cerebrovascular matrisome: A
convergent mechanism in small v essel disease of the brain?
J Cereb Blood Flow Metab. 2016 ; 36,143-157.
68. Comănescu M., Arsene D., Ardeleanu C., Bussolati G. The
mandate for a proper preservation in histopathological
tissues. Rom J Morphol Embryol. 2012 ; 53, 233-242.
69. García -Cabezas M.Á., J ohn Y.J., Barbas H., Zikopoulos B.
Distinction of Neurons, Glia and Endothelial Cells in the
Cerebral Cortex: An Algorithm Based on Cytological
Features. Front Neuroanat. 2016 ;10, 10.