Journal of Structural Biology 212 (2020) 107659

Contents lists available at ScienceDirect

Journal of Structural Biology

journal homepage: www.elsevier.com/locate/yjsbi

Investigation of the human pineal gland 3D organization by X-ray phase

contrast tomography
Inna Bukreeva a, b, *, Olga Junemann c, *, Alessia Cedola a, *, Francesca Palermo a, d,
Laura Maugeri a, e, Ginevra Begani Provinciali f, a, Nicola Pieroni a, g, Alessia Sanna a,
Dmitry A. Otlyga c, Alexey Buzmakov h, Yuri Krivonosov h, Denis Zolotov h, Marina Chukalina h, i,
Anna Ivanova h, Sergey Saveliev c, 1, Victor Asadchikov h, 1, Michela Fratini a, e, 1
a
Institute of Nanotechnology- CNR, Lecce Unit, Campus Ecotekne Via Monteroni, Lecce; Rome Unit, Piazzale Aldo Moro 5, Rome, Italy
b
P.N. Lebedev Physical Institute, RAS, Leninskiy pr., 53 Moscow, Russian Federation
c
FSSI Research Institute of Human Morphology, Tsyurupy Str 3, Moscow, Russian Federation
d
Department of Physics, University of Calabria, I-87036 Arcavacata di Rende (CS), Italy
e
IRCCS Fondazione Santa Lucia, Via Ardeatina 352, Rome, Italy
f
Laboratoire d’Optique appliquée, ENSTA Paris, Institut Polytechnique de Paris, 828 boulevard des Maréchaux, Palaiseau, France
g
SAIMLAL Department, Sapienza University, via A. Scarpa 14, Rome, Italy
h
FSRC «Crystallography and Photonics» RAS, Leninskiy pr., 59, Moscow, Russian Federation
i
Smart Engines Service LLC, 60-letiya Oktyabrya pr., 9, Moscow, Russian Federation

A R T I C L E I N F O A B S T R A C T

Keywords: Pineal gland (PG) is a part of the human brain epithalamus that plays an important role in sleep, circadian
X-ray phase contrast imaging rhythm, immunity, and reproduction. The calcium deposits and lesions in PG interfere with normal function of
X-ray micro-tomography the organ and can be associated with different health disorders including serious neurological diseases. At the
Human pineal gland
moment, the detailed mechanisms of PG calcifications and PG lesions formation as well as their involvement in
Pinealocytes
Pineal cysts
pathological processes are not fully understood. The deep and comprehensive study of the structure of the uncut
Pineal calcifications human PG with histological details, poses a stiff challenge to most imaging techniques, due to low spatial res­
olution, low visibility or to exceedingly aggressive sample preparation. Here, we investigate the whole uncut and
unstained human post-mortem PGs by X-ray phase contrast tomography (XPCT). XPCT is an advanced 3D im­
aging technique, that permits to study of both soft and calcified tissue of a sample at different scales: from the
whole organ to cell structure. In our research we simultaneously resolved 3D structure of parenchyma, vascular
network and calcifications. Moreover, we distinguished structural details of intact and degenerated PG tissue. We
discriminated calcifications with different structure, pinealocytes nuclei and the glial cells processes. All results
were validated by histology. Our research clear demonstrated that XPCT is a potential tool for the high resolution
3D imaging of PG morphological features. This technique opens a new perspective to investigate PG dysfunction
and understand the mechanisms of onset and progression of diseases involving the pineal gland.

1. Introduction nucleus (Erlich and Apuzzo, 1985). PG parenchyma is composed mainly

of pinealocytes secreting melatonin, microglia and astrocytes. Calcium
The pineal gland (PG) is part of the human brain epithalamus located deposits are a common occurrence in human PG. They are progressively
in the geometric center of the brain. This is a structurally complex accumulated in PG tissue with age and are not usually considered as
asymmetric formation with average size of 1 cm3 (Golan et al., 2002). pathology (Koshy and Vettivel, 2001; Schmid, 1993). Recent studies on
The PG is a central structure in the circadian system, which produces PG have shown that PG calcification can occur both outside and inside
melatonin under the control of the central clock and the suprachiasmatic cells. Calcification was observed in PG matrix, and in the nucleus of

* Corresponding authors at: Institute of Nanotechnology – CNR, Rome Unit, Piazzale Aldo Moro 5, Rome, Italy (I. Bukreeva).
E-mail addresses: [email protected] (I. Bukreeva), [email protected] (O. Junemann), [email protected] (A. Cedola).
1
Equal contribution.

https://doi.org/10.1016/j.jsb.2020.107659
Received 26 June 2020; Received in revised form 19 October 2020; Accepted 20 October 2020
Available online 24 October 2020
1047-8477/© 2020 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

pinealocytes (Kunz et al., 1999). Calcified tissue as well as cystic lesions prepared to study the morphology and structure of the organ. In Table 1
and a variety of neoplastic, congenital masses and tumors in PG may we have reported the description of the samples including the name of
interfere with normal PG function. Consecutively malfunctions of pa­ the sample, gender and age.
renchyma cells secreting melatonin might be associated with sleep dis­ The study was carried out on autopsy material obtained from the
orders, cancer, several psychiatric diseases, as well as collection of Federal State Scientific Institution Research Institute of
neurodegenerative diseases, e.g. Parkinson’s and Alzheimer’s disease Human Morphology (Moscow, Russian Federation). All protocols were
(Bruno et al., 2019; Kunz et al., 1999; Song, 2019; Tan et al., 2018). Up approved by the Ethical Committee of the Research Institute of Human
to the present time the detailed mechanisms of PG calcifications and PG Morphology of the Russian Academy of Medical Sciences (now FSSI
lesions formation are not fully understood and many questions about the Research Institute of Human Morphology) (No. 6A of October 19, 2009)
relationship between PG mass formation to neurosecretory cell pathol­ and are in correspondence with instructions of the Declaration of Hel­
ogies and vascular network malfunctions remain unanswered. sinki including points 7–10 for human material from 12.01.1996 with
Achieving a three-dimensional image of the whole human PG the last amendments from 19.12.2016.
(characterized by a high degree of anatomical complexity) comparable
in quality to histology poses a stiff challenge for most imaging tech­ 2.2. Samples preparation for micro-CT and XPCT
niques due to low spatial resolution, low visibility, or extremely
aggressive sample preparation methods. Autopsy material (the whole pineal glands PG-I, PG-II, PG-III and PG-
Thus, there is an increasing demand for a comprehensive study of the IV) was fixed in 10% formalin solution. Before performing the experi­
morphology of the whole PG at the level of cells and capillaries, using ments, the samples were placed and maintained in 70% ethanol.
non-destructive, high-resolution 3D imaging techniques. In particular,
the high-resolution visualization of degenerated tissues – inaccessible to 2.3. Micro-CT Set-up
standard 3D imaging techniques – would provide new opportunities for
the understanding of the formation mechanisms of PG calcifications and Micro-CT measurements were performed at TOMAS microtomo­
lesions. graphic laboratory setup, developed and operating at the Federal Sci­
X-ray phase contrast tomography (XPCT) is based on both effects – entific Research Centre “Crystallography and Photonics” of Russian
the attenuation and the phase shift of the x-ray beam transmitted Academy of Sciences, described in detail in (Buzmakov et al., 2018). A
through a sample. It is capable to detect features in high and low standard x-ray tube with a molybdenum anode was used as a source. The
absorbing biological tissue and analyze a sample at different scales: from values of the accelerating voltage and current were 40 kV and 20 mA,
whole-organ to cell structure. Recent publications (Bravin et al., 2013; respectively. The energy of the probing radiation was 17.5 keV (pyro­
Cedola et al., 2017; Fratini et al., 2015; Massimi et al., 2019) have graphite crystal was used as a monochromator). In each experiment, 400
demonstrated the ability of propagation-based imaging (PBI) to visu­ radiography projections were measured in an angular range of 200 de­
alize the 3D architecture of the central nervous system – the neuronal grees with a step of 0.5 degrees. The measurements were carried out in
and vascular networks, in particular – at micrometric and sub- parallel scanning scheme. Detector XIMEA xiRAY11 had a pixel size 9x9
micrometric scale. We applied PBI based XPCT as a relevant tool for micron, the total scan time was 120 min.
3D studies of uncut PG dissected from human post-mortem brain. PBI
set-up exploits x-ray propagation in free space to visualize the internal
2.4. XPCT Set-up
structure of the sample and achieves high spatial resolution with a large
field of view. Unlike standard imaging methods such as histology or CT,
XPCT measurements were performed using the Propagation Based
XPCT enables the high-resolution visualization of both x-ray transparent
Imaging (PBI) set up (Bravin et al., 2013). PBI exploits intensity varia­
brain tissue and x-ray absorbing calcification (Fratini et al., 2015;
tions via propagation of wave front between the object and the detector.
Khimchenko et al., 2018; Massimi et al., 2019; Pacilè et al., 2019;
Samples PG-I and PG-II were measured at the ID17 beamline (ESRF)
Töpperwien et al., 2020) without destructive sectioning and without the
using a pink beam. The peak energy was 60 keV. The tomography was
need for exogenous contrast agents.
acquired in half-acquisition mode (Wang, 2002) with 3045 projections
In this work, we studied the internal structure of four human PGs
and an exposure time of 0.4 s, covering a total angle range of 360 de­
with different extent of intrapineal calcifications and PG cystic lesion
grees. Samples were placed at 1.2 m from the imaging system (Mittone
using micro-CT, XPCT, and histology. Here we presented, to the best of
et al., 2017) with pixel size of 3.5 µm.
our knowledge, the first nondestructive high-resolution 3D investigation
PG-III and PG-IV were measured at the P05 beamline of the syn­
of PG morphology at different scales: from the whole organ to cell. We
chrotron facility PETRA III, DESY, operated by the Helmholtz-Zentrum
identified macro- and microscopically different tissues such as paren­
Geesthacht (PETRA III, DESY) (Wilde et al., 2016) using a mono­
chyma, vascular network, calcified tissue, and calcifications. XPCT
chromatic beam energy 25 keV. The tomography was acquired in half-
findings have shown the presence of a cystic lesion in PGs invisible in
acquisition mode (Wang, 2002) with 4000 projections and an expo­
micro-CT. Moreover, we observed the structural details of PG laminated
sure time of 0.25 s, covering a total angle range of 360 degrees. Samples
calcifications and features characterizing the degradation of the pineal
were placed at 0.5 m from the imaging system (Wilde et al., 2016) with
tissue in cystic PG. Specifically, we have distinguished the intact pa­
pixel size of 1.28 µm and 0.64 µm.
renchyma including pinealocytes and tissue with a fibrous structure
composed of glial cells processes replacing pinealocytes in cystic lesions.
2.5. Histology
Based on our research, we believe that XPCT imaging would be an
effective tool for a deep and comprehensive study of calcification and
The samples of human pineal glands PG-I, PG-II and PG-III were fixed
mass formation in PG and their relationship with pathologies and
vascular network malfunctions.
Table 1
Description of the samples.
2. Materials and methods
Samples Gender Age, years
2.1. Samples PG-I F 79
PG-II M 51
The post-mortem PG of four human subjects PG-I, PG-II, PG-III and PG-III F 70
PG-IV F 69
PG-IV without neurodegenerative diseases have been collected and

2
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

in 10% buffered formalin. Fixed tissues were dehydrated in eight por­ significant degree of calcification. In PG-III the intrapineal calcification
tions of absolute isopropyl alcohol IsoPrep (BioVitrum) and embedded was not detectable (the sample is not shown in the figure).
in paraffin blocks. Ten-micron sections were cut and mounted on glass In Table 2 we have reported the PGs morphological characteristics
slices for Mallory’s anillin blue connective tissue and Halocyanine obtained with micro-CT: pineal gland volume (PV), calcifications loca­
staining. tion (CaLoc), calcified volume (CaV), location of cysts (CyLoc), the total
volume of cystic lesions (CyV), and active pineal volume (APV = (PV-
2.6. Data processing and computational platform (CaV + CyV))/PV).
We classified the calcified deposits based on their localization (pineal
We performed flat- /dark-fields correction on raw data and each concrements localization CaLoc in Table 2). The main deposit of the
tomographic projection was normalized with the average value of concrements in PG-I was identified in the central part of the PG while the
background outside the object. The phase contrast retrieval algorithm concrements of PG-II were concentrated in the central part as much as in
(Paganin et al., 2002) was applied to each XPCT projection. After the the apical area of the epiphysis. Calcified deposits in PG-IV were iden­
phase retrieval procedure, we used the resulting set of data for 3D tified mainly in the central part and the apex of the sample. We per­
reconstruction of the object. Tomographic reconstruction was done with formed an evaluation of the PG volume (PV) and calcified volume (CaV)
Filtered Back Projection method (FBP). A post processing procedure has of the samples based on micro-CT. Volumes were segmented using
been applied to enhance visibility of calcifications and soft tissue of the global threshold binarization with a threshold of 0.4 mm− 1 for calcifi­
PG. Standard tools and plugins of the open-source image program cations and 0.078 mm− 1 for PG soft tissue. The estimated PV of PG-I, PG-
ImageJ/Fiji were used (Schindelin et al., 2012). In particular, average II, PG-III and PG-IV was about 246.7 mm3, 104.2 mm3, 98 mm3, and 180
intensity and maximum intensity projection options of the standard tool mm3, respectively. The estimated CaV of PG-I, PG-II and PG-IV was
“Z Project” of ImageJ/Fiji were used. Data pre-processing, artifact about 5.2 mm3, 13.2 mm3, and 11.5 mm3, respectively. The evaluation
removal, phase retrieval and FBP reconstruction were done with the shows a significant extent of calcifications in PG-II reducing the Active
open-source software toolkit SYRMEP Tomo Project (STP) (Brun et al., pineal volume (APV) by 13%.
2017; van Aarle et al., 2015). Maximum projection through the XPCT We noted that even if micro-CT imaging has provided high visibility
volume with a thickness 10 µm has been used for the comparison with of calcium deposits in the samples, however, detailed structure of the
the histology, which is a 2D technique and the information in depth of PGs soft tissue morphology including the cystic lesions could not be
histological section is averaged within 10 µm. Both maximum and distinguished.
average z-projections through XPCT volume with a different thickness
has been applied to highlight specific 3D features in the samples. 3.2. X-ray phase contrast tomography and histology in imaging of PGs

3. Results 3.2.1. Pineal gland concrements

Fig. 2(a) and Fig. 2(c) show XPCT images of the whole PG-I (a) and
X-ray micro-CT, XPCT and histological technique were used to study PG-II (c), respectively. The images are the result of average intensity
the morphology of three post-mortem human PGs: PG-I, PG-II, PG-III. projection of 50 slices, corresponding to 150-µm thick sample (see Data
and PG-IV. processing and Computational platform in Materials and Methods).
Gray-scale reconstructed slices of PG-I and PG-II are shown in Fig. 2S(a,
3.1. X-ray microtomography of PGs
Table 2
Micro-CT technique was used to select the samples in terms of Morphological characteristics of the samples: PV – pineal gland volume, Ca_Loc
different extent and size of pineal calcifications. Fig. 1(a) shows an – calcifications location, Ca_V – calcified volume, CyLoc – location of cysts, CyV
image of post-mortem human PG. Fig. 1(b) and 1(c) respectively display – total volume of cystic lesions, APV – active pineal volume.
the 3D image of PG-I and PG-II, obtained in micro-CT experiments. Gray- Samples PV, CaLoc CaV CyLoc CyV mm3 APV
scale reconstructed slices of PG-I and PG-II are shown in Fig. 1S(a-d) and mm3 mm3
1S(e-h) of Supplementary Materials (SM), respectively. Segmentation of PG-I 246.7 Center 5.5 Center 0.3 0.98
calcifications in PG-I and PG-II is shown in Fig. 1S(b,d) and 1S(f,h) of PG-II 104.2 Center, 13.2 Non Non 0.87
SM, respectively. apex detected detected
PG-III 98 Non Non Habenula 7.5 0.92
In Fig. 1(b,c) high absorbing calcifications are seen as white spots. detected defined region
Black spots, in Fig. 1(b,c), represent projections of the PG calcified area PG-IV 180 Center 11.5 Non Non 0.94
in three mutually orthogonal directions. In this regard, PG-I (Fig. 1(b)) apex detected defined
has a minor degree of calcification and PG-II (Fig. 1(c)) shows a

Fig. 1. (a) Photo of the human PG. (b) Micro-CT image of PG-I (c) Micro-CT image of PG-II Brain calcification is seeing as white spots. Projections of the PG calcified
area in three mutually orthogonal directions are shown in figures (b) and (c) as black spots.

3
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 2. (a) XPCT image of PG-I (b) Calcifica­

tion in the internal part of PG-I. (c) XPCT
image of PG-II. (d) Calcifications in the inter­
nal part of PG-II. (a-d) Images are obtained
using the average intensity through a slab of
50 slices, corresponding to 150-µm-thick sam­
ple. (e) XPCT image of the mulberry shaped
deposition in PG-II (white triple arrow). (f)
Cross-section of the mulberry shaped deposi­
tion shown in (e). (g) Histological section of
the concrement in PG-II, Mallory staining. (h)
XPCT image of PG-II with non-aggregated and
partially aggregated concrements (white dou­
ble arrow). (i) Histological section of PG-II
with calcifications, Mallory staining. (j) XPCT
image of small-sized calcifications in the septa
of PG-II (white dashed arrow). (k) Histology,
small-size calcifications scattered over the
septa of PG-II, Mallory staining. The thickness
of the slabs in XPCT and histological images is
about 10 µm.

b) and 2S(c,d) of SM, respectively. Major calcium deposits in both PG-I with laminar internal structure are typical for PG. Notably, the aggre­
and PG-II are concentrated in the PG’s matrix (intrapineal location) and gation of multiple concrements can lead to large-scale lamination on the
some diffused calculi are scattered in the pineal capsule (extrapineal whole aggregate (Kim et al., 2012). An accumulation of spherical non-
location). The zooms of PG-I and PG-II are shown in Fig. 2(b) and Fig. 2 aggregated and partially aggregated concrements with a bumpy shape
(d), respectively. PG-I contains mainly non-aggregated concrements (white double arrow) formed in PGII are shown in the XPCT image
with mulberry shaped appearance (white triple arrow in Fig. 2(b)) (Fig. 2(h)) and in the corresponding histological section (Fig. 2(i)).
concentrated in the central part of the PG and a few small-sized con­ Presumably, these friable concrements with their internal structure are
crements (white double arrow in Fig. 2(b)) scattered in the PG matrix. provisional and can be destroyed during life.
PG-II contains a large amount of aggregated (white calcified conglom­ The dense calcified conglomerates in PG-II, well seen in Fig. 2(c,d) as
erates visible in Fig. 2(c,d)) and non-aggregated concrements (white a massive white structure, and are the result of a multi-year’s fusion of
triple arrow in Fig. 2(d)) of different sizes and shapes. Non-aggregated smaller aggregates. Apparently, concrements of this type are the most
concrements were mainly detected in the periphery of PG-II. stable formations. The round shaped small-size mineral deposits scat­
A three-dimensional image of mulberry shaped concrement is shown tered over the septa of PG-II are shown in Fig. 2(j) (XPCT image, white
in Fig. 2(e). The concrement has scallop-shaped concentric laminations dashed arrow). The corresponding histological section is given in Fig. 2
in the internal structure, as shown in Fig. 2(f), which is in good agree­ (k).
ment with histological section (Mallory staining; magnification of mi­
croscope ×200) given in Fig. 2(g). In both Fig. 2(f) and Fig. 2(g) the 3.2.2. Pineal gland vascular network
thickness of sections is about 10 µm. The mulberry-like concrements PG is a highly vascularized organ with a profusion of arteries, veins,

4
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

and capillaries (Goldman and Wurtman, 1964). The blood vessels enter lumen. Notably, the connective tissue image in Fig. 4(a,b) has a high
the PG from the capsule via the connective tissue septa and form a dense contrast. This might indicate a high concentration of minerals in
network of interlobular capillaries, which drain into numeral pineal collagen that is consistent with the Ref. (Alcolado et al., 1986) where the
veins. Blood supply plays a fundamental role in PG melatonin secretion authors found calcification of the choroid plexus stroma in the region of
ability. calcified depositions formation.
Rich vascularization has been detected in PG-I within the septa Morphological analysis, performed with XPCT and validated by
around the lobules and in the interlobular connective tissue, inside and histology, detected a few blood vessels in the calcification-free part of
outside the calcified area. The volume image of PG-I with a dense PG-II. In the highly calcified part of the sample we observed some vessels
network of vessels segmented in red and calcifications segmented in with calcium deposits along the walls. The three-dimensional XPCT
white is shown in Fig. 3(a-c). Fig. 3 represents the axial (a) and longi­ images of a calcified vessel lumen is shown in Fig. 4(c). The longitudinal
tudinal (b,c) view of PG-I obtained as a maximum projection through the section of the vessel in Fig. 4(c) reveals the deposits attached to each
volume with a thickness of about 0.5 mm. Fig. 3(d) comprises zooms of other. The arrangement of the deposits is consistent with the shape of
Fig. 3(a-c) showing the intrapineal vascular architecture with the pro­ the vessel. The diameter of the vessel lumen is about 40 µm and the
fusion of PG blood vessels. Fig. 3(e-h) show XPCT (e,g) and histological average size of the deposits is less than 10–20 µm. The histological image
(h) sections of a branching PG vessel with the well-seen lumen and the given in Fig. 4(d) shows a longitudinal section of blood vessels scattered
vessel wall. The diameter of the vessel is about 50 µm. The thickness of with small concrements.
the wall is about 10 µm.
Fig. 4(a,b) illustrate the morphology of PG-II within the area of 3.2.3. Pineal gland soft tissue
calcification. In these figures, calcifications (marked with white arrows) The PG has a complex morphology. The parenchyma of the pineal
and fibrovascular stroma supporting the lobules (marked with yellow body composed of pinealocytes and neuroglial cells is surrounded by a
arrows) can be seen in white and are well distinguishable from the pa­ connective tissue capsule composed of pia mater. From the capsule, the
renchyma, which is shown in dark. Yellow dashed arrows in Fig. 4(b) septa pass into the PG and divides the parenchyma into incomplete
indicates the longitudinal section of PG trabeculae with well-seen lobules.

Fig. 3. A network of vessels revealed in PG-I. (a-g) XPCT image: (a) axial view and (b,c) longitudinal view of PG obtained as a maximum projection through the slab
with the thickness of about 0.5 mm, calcifications segmented in white and blood vessels segmented in red; (d) zooms of the intrapineal blood vessels network shown
in (a-c); (e-g) Images of the PG blood vessel with diameter of about 50 µm; (h) Histological section of the sample, Mallory staining. A branched vessel with uneven
blood filling is visible. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

5
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 4. XPCT image of PG-II. (a-b) Zoom of the central part: calcifications (marked with white arrows) and fibrovascular stroma supporting the lobules (marked with
yellow arrows) are seen in white. The longitudinal section of PG trabeculae is marked with yellow dashed arrows. (c) The lLongitudinal virtual cut of the blood
vessels scattered with calcium deposits. (d) Histological section of the blood vessel with lumen covered with calcium deposits, Mallory staining. (For interpretation of
the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5(a,b) and Fig. 5(c,d) provide zooms of the central part of PG-I 3.3. X-ray phase contrast tomography of PGs with high resolution
and PG-II, respectively, to better understand the organization of PG
soft tissue within the area of concrements formation. Fig. 5(e,f) are the In sample PG-III both XPCT and histology identified glial cysts with
zoom of the peripherical part of PG-II. The left column in Fig. 5 repre­ the gliotic area that was invisible in the micro-CT image. The XPCT axial
sents the lobular architecture of samples (lobules are marked with blue view of PG-III with the cystic lesion is shown in Fig. 7(a) (pixel size 1.28
arrows) obtained with XPCT imaging. The right column shows corre­ µm). Gray-scale reconstructed slice of PG-III and zoom of the cyst are
sponding histological sections images. In Fig. 5(a) fibrovascular stroma shown in Fig. 2S(f) of SM. The total volume of the cystic lesion is about
supporting the lobules of PG-I are visible in dark gray , while blood 7.5 mm3. Active pineal volume (APV) was reduced by 8%. (see Table 2).
vessels (marked with a red arrow) and small concrements are visible in To differentiate between the intact and degenerated PG tissue and
white. In Fig. 5(c,e) calcifications and fibrovascular stroma of PG-II (the increases the amount of information on the lesion morphology we used
stroma is marked with yellow arrows) are visible in white and well XPCT images with an enhanced spatial resolution (pixel size 0.64 µm)
distinct from the parenchyma visible in dark. The lobular structure in shown in Fig. 7(b,d) (a corresponding single slice of the sample is shown
the peripheral part of PG-II (Fig. 5e, f) is damaged, the interlobular septa in Fig. 3S(a) of SM). One can clearly distinguish in Fig. 7(b) between
are strongly thickened. intact PG parenchyma, where pinealocytes nuclei are visible as white
Fig. 6(a) displays the histological image of the PG with multiple dots, and the gliosis visible as a pinealocytes-less cyst wall composed of
fluid-filled cysts diagnosed post-mortem in PG-I. The glial cyst (green fibroblasts, glial cells and their processes. The histological section of the
arrow) with central coagulum and surrounding glial fibers (green double cyst (Mallory staining) is shown in Fig. 7(c). The structure of the gliotic
arrow) obtained with XPCT technique and histology are shown in Fig. 6 area is represented in Fig. 7(d) (XPCT image, a corresponding single slice
(b) and Fig. 6(c), respectively. The thickness of both the tomographic is shown in Fig. 3S(b) of SM) and Fig. 7(e) (histology, halocyanine
and the histological sections is about 10 µm (corresponding gray-scale staining). A good agreement between XPCT and histology results is
reconstructed slice is shown in Fig. 2S(e) of SM). In Fig. 6(b), small evident. Images in Fig. 7(b,d) were retrieved as the maximum intensity
blood vessels and intrapineal calcifications in area next to the cyst dis­ projection of 15 slices, through the slab corresponding to the 10-µm-
played as white spots cannot be distinguished from one another, while thick histological section.
3D image (thickness is about 200 µm) allows easy discrimination be­ The intrapineal vascular architecture of PG-III is shown in Fig. 8(a).
tween blood vessels (red arrows in Fig. 6(b, d)) and calcifications (white Pineal gland vasculature is visible in the figure as a network of large,
spots in Fig. 6(d)). Three cysts observed in PG-I had a total volume of small and fine blood vessels (diameter of vessels varies from 1 to 100
about 0.3 mm3. Both, the cystic lesion and calcifications reduced active µm). The image was retrieved as the maximum intensity projection
pineal volume (APV) of PG-I by 2% (see Table 2). through a 0.5 mm thick slab of the sample. Gray-scale reconstructed

6
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 5. PG parenchyma with the

lobular architecture of samples. (a, b)
The central part of PG-I: (a) XPCT im­
ages. (b) Histological section (Mallory
staining) corresponding to (a). The
lobular structure in (a, b) is partially
damaged, the interlobular septa in some
areas are strongly thickened. (c, d) The
central part of PG-II. (c) XPCT images.
(d) Histological section (Mallory stain­
ing) corresponding to (c). The lobular
structure in (c, d) is damaged, calcified
depositions are visible in the septa. (e, f)
Peripheral part of PG-II. (e) XPCT im­
ages. (d) Histological section (Mallory
staining) corresponding to (e). PG lob­
ules, connective tissue septa and blood
vessels are shown with blue, yellow and
red arrows respectively. (For interpre­
tation of the references to color in this
figure legend, the reader is referred to
the web version of this article.)

slices of PG parenchyma are shown in Fig. 3S(c,d) of SM. Fig. 8(b) PG calcifications are shown in Fig. 2S(g) and 2S(h) of Supplementary
represents XPCT image (10-micron-thick slab) illustarating PG paren­ Materials. In the sample both small separately lying concrements and
chyma with fibrovascular stroma (yellow arrow), blood vessels (red calcifications combined into large conglomerates were detected. In
arrow) and lobules with pinealocytes (blue arrow). Fig. 8(c) shows the Fig. 9a soft tissue and calcification are well distinguished. Fig. 9(b-g)
corresponding histological section (Mallory staining) that has a good illustrate the PG micro architecture with high resolution (pixel size 0.64
agreement with the tomography. Pineal gland cells are represented in µm). Fig. 9(b) and 9(c) show 3D image and tomographic slice image of a
Fig. 8(d). The pinealocytes in XPCT image are visible as a green cell’s separately lying concrement, respectively. The concentric laminations in
bodies with bright nuclei. The cells had an average diameter of about 9 the internal structure is well distinguished. A large conglomerate of
µm. Their nuclei had an average diameter of about 6 µm. Histological calcifications is shown in Fig. 9(d). The cross-section shown in Fig. 9(e)
image of cells obtained with halocyanine staining is shown in Fig. 8(e). reveals a noticeable large-scale lamination of the whole conglomerate.
The image of pinealocytes with processes surrounded by connective fi­ Three-dimensional image and tomographic slice of a large
bers is distinguishable in both Fig. 8(d) and Fig. 8(e). conglomerate within the soft tissue are shown in Fig. 9(f) and 9(g),
XPCT image of the sample PG-IV is shown in Fig. 9(a) (pixel size 1.28 respectively (corresponding tomographic slice and segmentation of PG
µm). Corresponding gray-scale reconstructed slice and segmentation of calcifications are shown in Fig. 3S(e) and 3S(f) of SM). The proliferation

7
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 6. Images of the cysts diagnosed

post-mortem in PG-I. (a) Histological
image (Mallory staining) of the PG
with multiple fluid-filled cysts. (b)
XPCT image of the cyst surrounded by
glial fibers, thickness of the slice is
about 10 µm. (c) Histological section
(Mallory staining) of the cyst, sur­
rounded by glial fibers. (d) XPCT
image of the PG with the cyst (thick­
ness of the slice is about 100 µm). The
glial cyst is marked with green arrows
while the glial fibers surrounding the
cyst are marked with green double
arrows, and small blood vessels are
marked with red arrows. (For inter­
pretation of the references to color in
this figure legend, the reader is
referred to the web version of this
article.)

of connective tissue in the parenchyma is noticeable. Pinealocytes nuclei histological quality the PG lobular structure and pinealocytes sur­
and blood vessel network are well seen. A network of capillaries with rounded by connective tissue spaces, PG vascular network and calcifi­
uneven blood filling is detected. cations of varying size embedded in a PG soft tissue. In addition, we
detected PG lesions and visualized the degenerated tissue fine structure
4. Discussion invisible in micro-CT. These results were confirmed by histological
examinations.
We used synchrotron-based free space propagation XPCT imaging to In PG-I we observed rich vascularization and a relatively small
perform 3D virtual histology on human PG. Unlike the standard histo­ quantity of PG concrements, which were mainly concentrated in the
logical method, XPCT allowed us to visualize the principal morpholog­ central part of the PG. On the one hand, both micro-CT and XPCT images
ical features of the whole PG without destructive sectioning of samples of PG-I have shown relatively small APV reduction in the sample due to
and exogenous contrast agents. We displayed in 3D with near- calcification. On the other hand, XPCT images revealed additional APV

8
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 7. Images of the cyst diagnosed post-mortem in PG-III. (a) XPCT image of the whole sample with a large cystic lesion. (b) XPCT image of the cyst surrounded by
glial tissue. (c) Histological section of the sample with the cyst, Mallory staining. (d) The structure of the glial tissue, XPCT image. (e) Glial tissue, histological section,
halocyanine staining. Thickness of both XPCT and histological sections is about 10 µm.

reduction due to multiple benign PG glial cysts in PG-I invisible in micro- analysis revealed a potentially high calcium concentration in the
CT. collagen of the fibrovascular stroma. We hypothesized that the vessels
In PG-II we observed extended intrapineal calcified depositions of and stroma pathology may be correlated with the formation of the
various sizes, scattered overall the organ. In particular, massive calci­ extended intrapineal calcifications in PG-II.
fications were observed in the center and in the apex of the PG, in cor­ In PG-III we observed two large glial cysts surrounded by a rim of a
respondence with histology. XPCT image analysis of PG-II within and glial tissue that reduced APV. Three-dimensional image, with a resolu­
outside the area of calcifications formation showed multiple calcified tion near to 1 µm, allowed clearly distinguishing between the intact PG
vessel lumens, both smooth walls and walls completely covered with parenchyma composed by pinealocytes and the pinealocytes-less area
calcified depositions were observed. The pathological appearance of the composed by fibroblasts, glial cells and their processes. By means of the
vessels suggests poor blood supply of the organ. Moreover, XPCT 3D image we also distinguished the pinealocytes structure including cell

9
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 8. Morphology of PG-III: (a) The

intrapineal vascular architecture of the
sample, XPCT image of 0.5 mm thick
slab. (b) PG lobular structure, the area of
parenchyma with fibrovascular stroma
(yellow arrow) blood vessels (red arrow)
and lobules with pinealocytes (blue
arrow), XPCT image of the 10-µm-thick
slab. (c) Histological section with PG
lobular structure, Mallory staining, ob­
jects and color of arrows corresponds to
(b). (d) XPCT image of pinealocytes
visible as a green cell’s bodies with
bright nuclei. (e) Histological image of
the pinealocytes with processes sur­
rounded by connective fibers, hal­
ocyanine staining. (For interpretation of
the references to color in this figure
legend, the reader is referred to the web
version of this article.)

bodies and nuclei. Besides, we segmented a network of PG microvessels. i.e. different incident energies (monochromatic and pink beam energy),
In PG-IV both, small separately lying concrements and calcifications different embedding material (air, agarose gel and paraffin) and
combined into large conglomerates were detected. Improved image different spatial resolutions (pixel from 3.5 µm to 0.64 µm), to attain an
resolution (pixel size 0.64 µm) allowed to distinguish the micro- acceptable quality of high-resolution 3D imaging of both PG soft tissue
architecture of the concrements with concentric lamination in the in­ and calcifications and their microstructure. We summarized the results
ternal structure. In large conglomerates of calcifications a noticeable as follows: incident monochromatic energy of about 25 provides good
large-scale lamination of the whole conglomerate was revealed. quality visualization of both soft tissue and concrements; the best results
However, we noted that the micron resolution was insufficient to were obtained in paraffin embedding material; one of the main chal­
resolve the features of neural and capillary networks that are important lenges in the 3D visualization of PG was image artifacts.
in the understanding of PG dysfunctions. Recent studies on PG have The critical issue to consider in the 3D imaging of PG is the artifacts
shown that PG calcification can occur both outside and inside cells generated by the strong-absorbing calcified concrements in the sample.
(Kunz et al., 1999). The problem of PG ultra-structure visualization can In some areas of the samples with extensive multiple calcifications, the
be solved with a sub-micron spatial resolution experiment that is the artifacts significantly degrade the quality of images. The problem may
next challenge to be dealt. be eventually overcome by the future improvement of computational
The samples were examined under different experimental conditions algorithms and the optimization of experimental methods and

10
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

Fig. 9. XPCT image of the sample PG-IV. (a) PG-IV tomographic slice (pixel size 1.28 µm). (b) 3D image of a calcification. (c) Tomographic slice of the calcification
with the concentric laminations in the internal structure. (d) Large conglomerate of calcifications. (e) Tomographic slice of the conglomerate with a noticeable
layered structure at the edges. (f) Large conglomerate of calcification within the soft tissue. (g) Tomographic slice of the soft tissue next to the conglomerate.

parameters. cysts with the histological details such as pinealocytes with their nuclei
in the parenchyma and the glial cells processes in the cystic glial tissue.
5. Conclusion
Funding
We believe that our research affords a new insight into studies on the
structures and functions of neuroendocrine PG while providing the high- Consiglio Nazionale delle Ricerche (“Accordo Bilaterale CNR/RFBR
quality and high-resolution 3D image of the whole uncut and unstained 2018–220” CUP B86C17000210005) & Russian Foundation for Basic
post-mortem organ. Progress in this field is currently hindered by a lack Research (18-52-7819); MIUR-CNR (“Tecnopolo diNanotecnologia e
of non-invasive high-resolution imaging techniques. We have shown Fotonica per la Medicina di Precisione”CUPB83B17000010001) &
that XPCT produces highly informative 3D visualization of PG soft and Regione Puglia (“Tecnomed” CUP B84I18000540002).
calcified tissue with near histological morphological features, which are
essential in the understanding of PG individual variability and PG pa­
thology. In particular, we have shown that XPCT allows nondestructive Declaration of Competing Interest
detection of calcifications and soft tissue lesions in PG. XPCT clearly
distinguishes between the intact PG tissue and the gliosis surrounding The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence

11
I. Bukreeva et al. Journal of Structural Biology 212 (2020) 107659

the work reported in this paper. multiple sclerosis model. Sci. Rep. 7, 1–11. https://doi.org/10.1038/s41598-017-
06251-7.
Erlich, S.S., Apuzzo, M.L., 1985. The pineal gland: anatomy, physiology, and clinical
Acknowledgements significance. J. Neurosurg. 63, 321–341. https://doi.org/10.3171/
jns.1985.63.3.0321 PM-2862230 M4.
Fratini, M., Bukreeva, I., Campi, G., Brun, F., Tromba, G., Modregger, P., Bucci, D.,
The bilateral project CNR/RFBR (2018- 2020) - accordo CNR-RFBR Battaglia, G., Spanò, R., Mastrogiacomo, M., Requardt, H., Giove, F., Bravin, A.,
delle Relazioni Internazionali (CUP B86C17000210005 & Russian Cedola, A., 2015. Simultaneous submicrometric 3D imaging of the micro-vascular
network and the neuronal system in a mouse spinal cord. Sci. Rep. 5, 8514. https://
number 18-52-7819), the national project RFBR (number 18-29-26028)-
doi.org/10.1038/srep08514 PM-25686728.
Russian Federation, the FISR Project “Tecnopolo di nanotecnologia e Golan, J., Torres, K., Staśkiewicz, G.J., Opielak, G., Maciejewski, R., 2002. Morphometric
fotonica per la medicina di precisione” (funded by MIUR/CNR, CUP parameters of the human pineal gland in relation to age, body weight and height.
B83B17000010001) and the TECNOMED project (funded by Regione Folia Morphol. (Warsz) 61, 111–113.
Goldman, H., Wurtman, R.J., 1964. Flow of blood to the pineal body of the rat. Nature
Puglia, CUP B84I18000540002) are acknowledged for financial support. 203, 87–88. https://doi.org/10.1038/203087a0.
We also thank the Ministry of Science and Higher Education of Russian Khimchenko, A., Bikis, C., Pacureanu, A., Hieber, S.E., Thalmann, P., Deyhle, H.,
Federation within the State assignment FSRC «Crystallography and Schweighauser, G., Hench, J., Frank, S., Müller-Gerbl, M., Schulz, G., Cloetens, P.,
Müller, B., 2018. Hard X-ray nanoholotomography: large-scale, label-free, 3D
Photonics» RAS for financial supporting and for the use of equipment of neuroimaging beyond optical limit. Adv. Sci. 5, 1700694. https://doi.org/10.1002/
the Shared Research Center “Structural diagnostics of materials” (proj­ advs.201700694.
ect RFMEFI62119X0035). M. Fratini acknowledges the Italian Ministry Kim, J., Kim, H.W., Chang, S., Kim, J.W., Je, J.H., Rhyu, I.J., 2012. Growth patterns for
acervuli in human pineal gland. Sci. Rep. 2 (984), 1–6. https://doi.org/10.1038/
of Health Young Researcher Grant 2013 (GR-2013-02358177) for srep00984.
financial support. Koshy, S., Vettivel, S.K., 2001. Varying appearances of calcification in human pineal
The authors thank the staff of the P05 beamline Fabian Wilde and gland: a light microscopic study. J. Anat. Soc. India 50, 17–18.
Kunz, D., Schmitz, S., Mahlberg, R., Mohr, A., Stöter, C., Wolf, K.J., Herrmann, W.M.,
Elena Longo at the P05 beamline of the synchrotron facility PETRA III,
1999. A new concept for melatonin deficit: On pineal calcification and melatonin
DESY, operated by the Helmholtz-Zentrum Geesthacht and Alberto excretion. Neuropsychopharmacology 21, 765–772. https://doi.org/10.1016/
Bravin at ID17 at ESRF for their experimental support. S0893-133X(99)00069-X.
Massimi, L., Bukreeva, I., Santamaria, G., Fratini, M., Corbelli, A., Brun, F., Fumagalli, S.,
Maugeri, L., Pacureanu, A., Cloetens, P., Pieroni, N., Fiordaliso, F., Forloni, G.,
Authors contributions Uccelli, A., Kerlero de Rosbo, N., Balducci, C., Cedola, A., 2019. Exploring
Alzheimer’s disease mouse brain through X-ray phase contrast tomography: From
the cell to the organ. Neuroimage 184, 490–495. https://doi.org/10.1016/j.
I.B., O.J., M.F., A.C., A.I.,V.A., S. S. wrote the main manuscript text; neuroimage.2018.09.044.
M.F., I.B., F.P., L.M, G.B.P., N.P. A.S., A.B., Yu..K., D.Z., O.J. carried out Mittone, A., Manakov, I., Broche, L., Jarnias, C., Coan, P., Bravin, A., 2017.
XPCT experiment, A.B. and Yu.K. carried out micro-CT experiment, I.B. Characterization of a sCMOS-based high-resolution imaging system. J. Synchrotron
Radiat. 24, 1226–1236. https://doi.org/10.1107/S160057751701222X.
and M.F. performed XPCT data processing, I.B. performed XPCT image Pacilè, S., Dullin, C., Baran, P., Tonutti, M., Perske, C., Fischer, U., Albers, J., Arfelli, F.,
processing, A.B., Yu.K., A.I., M.C. performed micro-CT data and image Dreossi, D., Pavlov, K., Maksimenko, A., Mayo, S.C., Nesterets, Y.I., Taba, S.T.,
processing and analysis, O.J., D.A.O., S.S. performed the histological Lewis, S., Brennan, P.C., Gureyev, T.E., Tromba, G., Wienbeck, S., 2019. Free
propagation phase-contrast breast CT provides higher image quality than cone-beam
examination and provided histological images of PGs, O.J., S.S. per­ breast-CT at low radiation doses: a feasibility study on human mastectomies. Sci.
formed micro-CT, XPCT image analysis. All authors reviewed the Rep. 9, 1–7. https://doi.org/10.1038/s41598-019-50075-6.
manuscript. Paganin, D., Mayo, S.C., Gureyev, T.E., Miller, P.R., Wilkins, S.W., 2002. Simultaneous
phase and amplitude extraction from a single defocused image of a homogeneous
object. J. Microsc. 206, 33–40. https://doi.org/10.1046/j.1365-2818.2002.01010.x
Appendix A. Supplementary data PM-12000561.
Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T.,
Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J.-Y., White, D.J.,
Supplementary data to this article can be found online at https://doi. Hartenstein, V., Eliceiri, K., Tomancak, P., Cardona, A., 2012. Fiji: an open-source
org/10.1016/j.jsb.2020.107659. platform for biological-image analysis. Nat. Methods 9, 676–682. https://doi.org/
10.1038/nmeth.2019 PM-22743772.
Schmid, H.A., 1993. Decreased melatonin biosynthesis, calcium flux, pineal gland
References calcification and aging: A hypothetical framework. Gerontology. https://doi.org/
10.1159/000213533.
Alcolado, J.C., Moore, I.E., Weller, R.O., 1986. Calcification in the human choroid Song, J., 2019. Pineal gland dysfunction in Alzheimer’s disease: Relationship with the
plexus, meningiomas and pineal gland. Neuropathol. Appl. Neurobiol. 12, 235–250. immune-pineal axis, sleep disturbance, and neurogenesis. Mol. Neurodegener.
Bravin, A., Coan, P., Suortti, P., 2013. X-ray phase-contrast imaging: from pre-clinical https://doi.org/10.1186/s13024-019-0330-8.
applications towards clinics. Phys. Med. Biol. 58, R1–R35. https://doi.org/10.1088/ Tan, D., Xu, B., Zhou, X., Reiter, R., 2018. Pineal calcification, melatonin production,
0031-9155/58/1/R1 PM-23220766. aging, associated health consequences and rejuvenation of the pineal gland.
Brun, F., Massimi, L., Fratini, M., Dreossi, D., Billé, F., Accardo, A., Pugliese, R., Molecules 23, 301. https://doi.org/10.3390/molecules23020301.
Cedola, A., 2017. SYRMEP Tomo Project: a graphical user interface for customizing Töpperwien, M., van der Meer, F., Stadelmann, C., Salditt, T., 2020. Correlative x-ray
CT reconstruction workflows. Adv. Struct. Chem. Imaging 3, 4. https://doi.org/ phase-contrast tomography and histology of human brain tissue affected by
10.1186/s40679-016-0036-8. Alzheimer’s disease. Neuroimage 210, 116523. https://doi.org/10.1016/j.
Bruno, F., Arrigoni, F., Maggialetti, N., Natella, R., Reginelli, A., Di Cesare, E., neuroimage.2020.116523.
Brunese, L., Giovagnoni, A., Masciocchi, C., Splendiani, A., Barile, A., 2019. van Aarle, W., Palenstijn, W.J., Beenhouwer, J., Altantzis, T., Bals, S., Batenburg, K.J.,
Neuroimaging in emergency: A review of possible role of pineal gland disease. Gland Sijbers, J., 2015. The ASTRA Toolbox: A platform for advanced algorithm
Surg. https://doi.org/10.21037/gs.2019.01.02. development in electron tomography. Ultramicroscopy 157, 35–47. https://doi.org/
Buzmakov, A.V., Asadchikov, V.E., Zolotov, D.A., Roshchin, B.S., Dymshits, Y.M., 10.1016/j.ultramic.2015.05.002 PM-26057688.
Shishkov, V.A., Chukalina, M.V., Ingacheva, A.S., Ichalova, D.E., Krivonosov, Y.S., Wang, G., 2002. X-ray micro-CT with a displaced detector array. Med. Phys. 29,
Dyachkova, I.G., Balzer, M., Castele, M., Chilingaryan, S., Kopmann, A., 2018. 1634–1636. https://doi.org/10.1118/1.1489043.
Laboratory microtomographs: design and data processing algorithms. Crystallogr. Wilde, F., Ogurreck, M., Greving, I., Hammel, J.U., Beckmann, F., Hipp, A.,
Reports 63, 1057–1061. https://doi.org/10.1134/S106377451806007X M4. Lottermoser, L., Khokhriakov, I., Lytaev, P., Dose, T., Burmester, H., Müller, M.,
Cedola, A., Bravin, A., Bukreeva, I., Fratini, M., Pacureanu, A., Mittone, A., Massimi, L., Schreyer, A., 2016. Micro-CT at the imaging beamline P05 at PETRA III. In: AIP
Cloetens, P., Coan, P., Campi, G., Spanò, R., Brun, F., Grigoryev, V., Petrosino, V., Conference Proceedings. American Institute of Physics Inc., p. 030035 https://doi.
Venturi, C., Mastrogiacomo, M., Kerlero De Rosbo, N., Uccelli, A., 2017. X-Ray phase org/10.1063/1.4952858.
contrast tomography reveals early vascular alterations and neuronal loss in a

12