Available via license: CC BY 4.0
Content may be subject to copyright.
Polyploid Monolayer Ishikawa Endometrial Cells form Unicellular
Hollow Spheroids Capable of Migration
Honoree Fleming, Ph.D.
Castleton State College, Castleton Vt., Dean of
Education, retired
founder CancerCellsinVitro.com
Castleton, Vermont
USA
Emails:
honoree.fleming@castleton.edu
Abstract
The results in this paper demonstrate that Ishikawa endometrial monolayer cells become
multinucleated by a process of nuclear “donation” from neighboring cells. As the resulting
polyploid cell detaches from the colony in which it was formed, it is possible to detect
mitonucleon(s) in the center of the cell. The mitonucleon is a transient mitochondrial
superstructure surrounding aggregated chromatin (Fleming et al. 1998) with characteristics of
the family of mitochondrial superstructures that are sometimes called spheroids or
cup-shaped mitochondria (Fleming, 2016a). As was recently demonstrated gas vacuoles form
within mitonucleons (Fleming, 2018). In the free-floating single cell, the retained gas creates
a central vacuole, and the cell becomes a spheroid that floats above the monolayer. It
resembles a “signet ring cell” in being characterized by a central vacuole and chromatin
compressed against the vacuole membrane. The resulting structure is a spheroids that is
hollow and unicellular, albeit polyploid. But whereas signet ring cells are assumed to be
undergoing apoptosis, that is not the case for unicellular spheroids. Complete spheres with
chromatin and cytosolic cell contents compressed against the cell membrane can be found
floating independently above Ishikawa monolayers. When an isolated sphere settles back
onto the surface of the petri dish, it is possible to observe dissipating gas bubbles within the
now flattened sphere for a short period of time. When the gas is discharged the resulting cell
looks like a typical giant polyploid cell.
Introduction
Almost 20 years ago, we described a transient mitochondrial superstructure formed when
confluent Ishikawa (Nishida et. al., 1973) endometrial cells were stimulated to differentiate
into fluid-filled hemispheres (Fleming et al., 1998; Fleming, 2016a,b,c). These hemispheres
can ultimately extend into gland-like structures, a frequent epithelial differentiation that
takes place in organs throughout the body and in pre-menopausal women every month in the
endometrial lining of the uterus.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Within the first few hours following monolayer stimulation, large structures staining for
endogenous mitochondrial carboxylases can be seen lying next to nuclear aggregates in
developing syncytia (Fleming, 2018). The resulting spheroidal superstructure which
eventually completely envelops aggregated nuclei fosters vacuole formation. Small vacuoles
form within the chromatin and a second much larger vacuole forms, contiguous with, but
separate from, the enclosed chromatin mass. As the vacuole expands within the
mitochondrial superstructure, it causes a significant protrusion of the syncytial apical
membrane against which it presses the enveloped chromatin.
We have called this structure a mitonucleon, hoping that the name communicates the essence
of this transient mitochondrial superstructure surrounding aggregated chromatin. On the
basis of this structure, It seems reasonable to conclude that the mitonucleon belongs to a
family of mitochondrial superstructures that others have called spheroids or cup-shaped giant
mitochondria that can also surround chromatin. There are particularly relevant similarities
between the mitonucleon (Fleming, 2018) and the mitochondrial superstructure that has
been called nebenkern and is responsible for the elongation of grasshopper (and other insects)
spermatid tails (Fleming, 2018) This paper describes the observations that in addition to
forming in syncytia mitonucleons can develop in a single Ishikawa endometrial monolayer cell
if it becomes multinucleated.
Results
The structure in fig. 1a is larger than the surrounding cells and stains intensely for chromatin
using hematoxylin and eosin. Cells immediately around it are oriented toward, sometimes
even appearing to be physically in contact with, this enlarged central cell. That orientation
together with the density of staining in the receiving cell suggests nuclei are being donated
from surrounding cells into the central cell. Although this is not a phenomenon frequently or
perhaps even ever previously recorded in cell culture, the transfer of “nurse cell contents”
into germ cells was observed decades ago in drosophila (Painter, 1959).
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Fig. 1a Enlarged cell staining both for chromatin
(dark central mass) and for biotin (pink material
surrounding nuclei).
Bar=50 microns
Fig. 1b Enlarged cell and neighboring syncytium.
Faint pink stains from biotin are seen in both
structures. This polyploid single cell was found in
a dish being induced for dome formation. Such
cells are also found in cultures that are not being
induced. Bar=50 microns
As fig. 1b shows, this enlarged multinucleated cell was found close to a syncytium that is also
multinucleated. Fig. 1b demonstrates that two different mechanisms can result in two very
different looking polyploid structures in Ishikawa cell cultures. The syncytium in fig. 1b
covers an area as large as 50 or more cells following fusion of a similar approximate number
of mononucleated cells. Nuclei in the syncytium appear to be fading, possibly signalling a
chemical change in chromatin as they begin to fuse into aggregates. The neighboring
multinucleated cell is far more compact, but still at least four to five times larger than
monolayer cells. There is no visible border around the syncytium other than what is formed
by neighboring mononucleated monolayer cells. The multinucleated cell in fig. 1a has a
distinct cell membrane and some of the neighboring cells are fusiform with one polar end
appearing to be in contact with the enlarged cell.
Staining only for endogenous biotin which we and others have shown is bound to
mitochondrial carboxylases (Fleming et.al., 1998; Gamache et. al., 2003 ), fig 2a reveals two
additional characteristics of an enlarged acceptor cell and the surrounding donor cells. So
much biotin stain is taken up by the receiving cell in fig. 2a that it is not possible to
distinguish much else within the cell suggesting that, in addition to multiple nuclei, large
amounts of endogenous biotin in mitochondria are also donated to the receiver cell. In the
surrounding cells the endogenous biotin is in the cytosol highlighting the extreme fusiform
shape of the cells oriented toward the “accepting” cell. Nuclei with prominent nucleoli in the
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
five cells making contact with the receiving cell between 10 o’clock and 2 o’clock appear to
have been fixed in the process of moving into the central cell. Nuclei cannot be seen in the
cells on the lower right side of the receiving cell and these cells appear to be rounding up,
suggesting that they have already donated most of their contents to the receiving cell.
Fig. 2b shows a multinucleated cell in a living culture. The donor cells in this figure are
slightly out of focus suggesting that the central cell has not only increased in size in the plane
of the dish, but has also begun to swell. One or two fusiform cells can be seen at 9 o’clock.
Multiple nuclei are clearly visible within the receiving cell.
Fig. 2a In this culture stained for biotin, it is
almost impossible to learn anything about the
contents of the receiver cell except that it
contains enough stainable biotin to obscure
anything else in the cell. Cells that appear to be
donors seem to comprise a couple of populations.
Nuclei in the six cells from ten o’clock to two
o’clock appear to be moving into the cell along
with biotin staining material from the cytoplasm.
The cell as 3 o’clock does not appear to be
attached and a nucleus is not visible. It’s
possible that the cells at 5 and 6 o’clock are still
attached.
Bar=50 microns
Fig. 2b Without the biotin stain, it is clear that
the central cell is filled with nuclei. In both
examples, cells around the central cell are not as
dense as in the rest of the monolayer,
presumably because they detach, following the
donation of cell contents.
Bar = 50 microns
It is particularly obvious in figs. 2 a and b that cells immediately surrounding the enlarged cell
are thinning out, a not surprising development if, as the pictures indicate, they have donated
nuclei and cytosol into the large central cell. Nurse cells in drosophila oocyte are also known
to “die off.”(Painter, 1959)
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
An enlarged cell staining for chromatin in the center with endogenous biotin out to the edges
of the cell is shown in fig. 3. Almost all of the surrounding cells have detached. Several
processes extend from the multinucleated cell back to the colony.
Fig. 3
Polyploid cell rounding up and detaching from an
Ishikawa monolayer. The cells stains intensely
for chromatin in the center surrounded by
endogenous biotin. Several processes extend
from the cell back into the colony.
Bar=50 microns
In fig 4 an enlarged cell above the monolayer, fixed and stained for endogenous biotin, shows
biotin-staining structures in the center of the detaching enlarged cell. These structures look
like mitonucleons initially seen in Ishikawa syncytia developing into domes. (Fleming et al.
1995; Fleming, 2016)
Fig.4 Detaching cell containing mitonucleons.
The biotin stain has moved to the center of the
cell. Mitonucleons appear in the center of
syncytia of differentiating Ishikawa cells during
dome formation. (Fleming et al. 1998; Fleming,
2016a)
Bar=50 microns
The mitonucleons are elaborated when multiple giant mitochondria become dense around
aggregated nuclei and then round up into spheroidal structures enveloping chromatin . Within
these structures, the aggregated nuclei become pycnotic as a result of gas vacuoles forming
within the mitonucleon. (Fleming, 2018) Pycnotic chromatin is assumed to be symptomatic
of apoptosis. But in dome differentiation, and again in the formation of hollow spheroids,
pycnosis does not signal cell death. Retained gas results in a swelling of the cell into a
floating spheroid that can be found in the medium above the monolayer (fig. 5).
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Fig. 5
Spheroid structures staining for endogenous
medium are found in medium above monolayers.
As gas vacuoles enlarge within one or more
mitonucleons within single cells, the endogenous
biotin stain can be detected. Two spheroids at
different stages of expansion can be seen in this
field above the out of focus monolayer. The
smallest structure at 2 o’clock may be the
earliest stage of a spheroid detaching from the
monolayer.
Bar=50 microns
There appear to be at least 3 structures in this figure that stain for biotin. The smallest, at 2
o’clock is mostly out of focus and may still be attached to the monolayer. The second
spheroid at 12 o’clock is clearly detached from the monolayer with a third spheroid nearby. In
this spheroid something like a window can be detected in the midst of the biotin stain.
Looking at a couple of similarly large spheroids not stained for biotin in Fig. 6, suggests that
this “window” reflects some compartmentalization as cellular material is being pressured
against the cell wall of the spheroid by a growing gas vacuole.
It is possible to detect two compartments in the unstained enlarging spheroids in fig. 6. One
compartment appears to be empty and is, all evidence suggests, filling with gases retained in
the mitonucleon(s). Those gases are compressing the cellular material of the other
compartment against the wall of the spheroid. Cytoplasm and structures resembling
compressed, giant nuclei are being compressed against the cell wall as the mitochondrial
superstructure generates and retains gases.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Fig. 6a
Unstained spheroid provides a glimpse of what
happens to cellular material as the spheroid
vacuole enlarges. Approximately two thirds of
the spheroid appears to be empty. The results
suggest that this enlarging vacuole flattens cell
material as well as nuclei against the membrane
of the expanding spheroid.
Bar=25 microns
Fig. 6b
Another unstained spheroid in the process of
forming a unicellular hollow spheroid.
Bar=50 microns
The pressure exerted by the enlarging gas vacuole eventually results in the structure shown
in figs. 7a and 7b, pictures taken of a floating hollow unicellular spheroid focusing first on
nuclei and then on one polar end of the spheroid. When the objective of the microscope is
focused on two giant nuclei it can be seen that they exist within a rim of cytoplasmic material
stretching around the spheroid. Focusing up, the nuclei are no longer visible at one polar
edge of the spheroid, although a dark circular structure can be seen, presumably within that
thin layer of cytoplasm detectable in fig. 7a.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Fig.7a Polyploid unicellular hollow spheroid.
Objective focused on giant nuclei.
Bar=25 microns
Fig.7b Polyploid unicellular hollow spheroid.
Objective focused on polar end of spheroid.
Bar=25 microns
Cellular spheroids were carefully collected from the petri dish and transferred into another
dish to which serum was added. Structures like that in fig. 7 would be expected to be
delicate, but inspection of the dish indicated that multiple spheroids survived the transfer.
Under these conditions unicellular hollow spheroids reattach to the surface of the petri dish,
releasing some kind of gas or mixture of trapped gases. These bubbles assumed to be
trapped in the transient mitonucleon structure dissipate within minutes leaving behind giant
cells containing one or more enlarged nuclei.
Fig. 8a. Central vacuole of hollow
Ishikawa spheroids contain gases. When
serum is added to medium containing
hollow spheroids, the floating spheroids
Fig. 8b. Bubbles dissipate
rapidly.
Bar=25 microns
Fig. 8c Multiple giant polyploid
cells such as this cell plated
down serum-free medium
containing hollow spheroids was
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
“pancake” onto the surface of the petri
dish. For a short time after the
initiation of reattachment, it is possible
to observe gaseous bubbles within the
structures.
Bar=25microns
transferred to a culture dish
with the addition of serum.
Bar=25 microns
The colonies in figs 8a, b, and c provide additional evidence that the enlargement of
multinucleated cells into hollow spheroids occurs because gas is trapped within these
structures.
Hollow spheroids are also sometimes observed attached to solid spheroids as shown in Figs. 9a
and 9b. In each case, the microscope is focused on the most prominent structure in the
hollow spheroid, one or more large nuclei, presumably polyploid, flattened against the outer
cell membrane.
Fig. 9a Hollow spheroid developing apparently
attached to a solid spheroid. Based on what is
seen in monolayer and shown in this paper, best
guess is that a cell associated with the solid
spheroid becomes multinucleated, develops
mitonucleons that retain gasses.
bar=50 microns
Fig. 9b Another example of a hollow spheroid
apparently attached to a solid spheroid.
bar=50microns
When serum is added to medium containing structures like these, spheroid cells revert to
monolayers over time. When the solid spheroid also has an attached hollow spheroid, part of
the resulting colony presents with the distinctive look of giant nuclei packed together in a
common pool of cytoplasm. In fig. 10a, the edges of the extension suggest a splattering of
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
the outer membrane of the “cell” as the hollow spheroid became attached rather like a
“burst” balloon. It is expected that giant nuclei such as those in fig. 10a will bud out, a
process demonstrated for giant nuclei in a variety of systems (Zybina &Zybina, 2002;
Erenpreisa et al. , Walen, Sundaram et al. ) the previously hollow spheroid takes on a
distinctive look, different from the rest of the colony with smaller nuclei presumably budding
off giant nuclei within an abundance of cytoplasm.
Fig. 10a Just as hollow spheroids can plate down
under appropriate conditions an attached hollow
spheroid can plate down. The scalloped edge of
the structure suggests some internal pressure in
what was once a hollow spheroid resulted in a
membrane blow-out.
Bar=25 microns
Fig. 10b The absence of distinct cell membranes
around the 8 or more nuclei in the extension of
this colony suggest that these nuclei budded off
from polyploid nuclei associated with a hollow
spheroid outgrowth of a solid spheroid. It is
worth noting that the nuclei that have budded
off are not uniform in size suggesting nuclei with
different n values just as has been described in
other systems.
Bar=50 microns
Discussion
Polyploid cells, particularly those in which multiple nuclei have fused into one or more giant
nuclei such as are seen in figs. 8 and 10, were once thought unable to produce progeny. But in
the past couple of decades, polyploid cells have emerged as more than simply diploid cells
gone bad. There is growing interest in how they arise and what their functions might be since
a number of reports have clearly demonstrated that they can produce progeny. Studying the
naturally occurring polyploid trophoblasts for more than two decades, Zybina and Zybina
reviewed that work in a 2008 paper discussing the fragmentation of giant nuclei into smaller
nuclear fragments and the budding of these into cytoplasm. Erenpreisa and colleagues (2000)
demonstrated a similar phenomenon in cells subjected to irradiation that induces mitotic
catastrophe. Most cells die, but some cells become polyploid, synthesizing DNA, but not
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
dividing. Erenpreisa showed that giant polyploid cells surviving this insult are able to give
rise to mitotically capable diploid cells in a process both she and Zybina call
depolyploidization. Budding is another term that has been used to describe the manner in
which polyploid cells give rise to diploid cells. Walen (2002) described the budding off of
normal diploid cells from human cells rendered polyploid by viral transformation (2002) as
Sundaram et al. (2004) did for polyploid cells that arise when mouse embryo fibroblast lines are
exposed to carcinogens,
The consensus seems to be that endocycling, DNA replication without cytokinesis, is the
predominant way in which polyploid cells arise, with syncytial formation recognized as an
alternative mechanism. It is syncytial formation that gives rise to polyploid predomes in
Ishikawa endometrial cells as previously described (Fleming 1998 and Fleming 2016).
In this paper we have shown that monolayer cells can become multinucleated by a third
mechanism not previously described in cell monolayers. It appears that fewer than 1% of
monolayer cells become polyploid because surrounding cells donate nuclei and cytosol
contents into a central enlarging cell. It is not clear what signals within the culture bring
about the coordination of cells involved in this process. That it happens is suggested by the
presence of fusiform cells surrounding, and in some instances appearing to be attached to, an
enlarged polyploid cell. Accessory cells such as these have been called “nurse cells” in other
systems. A well- known example involves drosophila melanogaster whose egg chamber
contains 16 cells. One cell will become the oocyte, the other 15 cells, the nurse cells, will
donate organelles, proteins and nucleic acids to the oocyte, preparing the egg to support the
development of an embryo until sustenance from the environment becomes available
(Painter, 1959).
Polyploid receiving cells far along in this process are sometimes found with few, if any,
surrounding cells presumably because donor cells die following transfer of cell contents. As
the polyploid cell itself detaches, multiple cell processes can be detected reaching back into
the colony from such cells. Processes back to the cell colony can be detected even as the
polyploid cell begins to become elevated, resulting for a time in “tethered” detached cells. Such
processes may be similar to cytonemes that are believed to facilitate communication (Kornberg
& Sugata R., 2014). Multiple nuclei occupy the center of the cell surrounded by material
staining for endogenous biotin.
In the final stages of detachment the biotin stain is found in the center of polyploid cells in
structures that resemble the mitochondrial superstructures (mitonucleons) that envelop
nuclear aggregates during dome formation (Fleming et al. 1998; Fleming 2016; Fleming 2018).
One function of these mitonucleons appears to be the retention of gases (Fleming 2016;
Fleming, 2018), and in a single detached cell the retention of gases results in the formation of
a unicellular, albeit polyploid, floating spheroid.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Spheroids of various sizes can usually only be seen by focusing above the monolayer. When
such spheroids are not stained for biotin the internal structure of hollow spheroids reveals
cellular material in the process of being compressed against the cellular membrane.
The results in this paper also show that the gases building up exert pressure resulting in cell
contents, including nuclei, becoming marginalized, flattened between the mitonucleon and
the outer cell wall (fig. 7) forming unicellular hollow spheroids.The gases at the center of
such a structure “become visible” as trapped bubbles in the minutes after hollow spheroids
attach to a petri dish in response to the addition of serum. Gas bubbles, initially visible,
rapidly disappear.
What cannot be said for certain is the identity of the gases that are building up. Carbon
dioxide is a common by product of aerobic metabolism but it has long been an article of faith
that this gas mostly passes freely in and out of cells. In fact however, variability in the rate
of diffusion as has been demonstrated in experiments in gastric gland cells (Waisbren et al.
1994). Endevard and colleagues (2013, 2014) have demonstrated a lower rate of diffusion of
CO2 through the apical as compared to the basal membrane of the guinea pig colon.
Furthermore, channels have been shown to be involved in CO2 transport under certain
conditions (Nakhoue et al. 1998; Itel et al. 2012). Is the passage of CO2, the most abundant
by-product of mitochondrial metabolism slowed by the mitochondrial superstructure?
Interestingly enough Armstrong and his colleagues found giant spheroidal mitochondria in
human endometrium at around the time of ovulation, when gland formation begins. This is a
finding that suggests what we are studying in vitro
is relevant in vivo
. Using an electron
microscope, they were able to show that these structures were surrounded a single sac of
endoplasmic reticulum. (Armstrong et. al, 1973) Might this unusual presence of an additional
membrane explain a build up of gases?
What about other gases known to evolve as a result of metabolism and to be active as gaseous
transmitters. Most of these appear to be produced in small amounts and to have a short half
life, although half life in a mixture of gases might be different. Perhaps the most intriguing
possibility, and purely speculative, is that mitochondrial superstructures bring about changes
in the mitochondria that result in a temporary anaerobic environment in which H2 can be
generated. Embley and colleagues (2003) have published a paper speculating that
mitochondria and hydrogenosomes are two forms of the same fundamental organelle. And,
Boxma et al. (2005) have demonstrated the existence of anaerobic mitochondria that produce
hydrogen. Admittedly speculative does a fraction of the total mitochondria in endometrial
epithelia forming domes or spheroids “revert” and begin producing hydrogen, a gas that
would certainly give spheroids a lift! (Fleming, 2018; Fleming, 2016a)
In this regard, it is important to note that the mitonucleon is transient. Mitonucleons active
in Ishikawa differentiation result in the accumulation of gases that unfurl the apical
membrane and compress aggregated chromatin against that membrane. The structure
becomes distended with gas and subsequently falls apart. The evidence in this paper suggests
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
that the component parts of the mitonucleon membrane are absorbed back into the giant cell
that forms when the spheroid reverts to a monolayer cell. Furthermore, the evidence,
particularly for mitonucleons observed during dome formation, is that only a fraction of
mitochondria might be generating hydrogen within the mitonucleon, while the remainder of
the mitochondria out in the cytoplasm, not confined within the mitochondrial superstructure,
are expected to remain capable of aerobic metabolism.
Multicellular Hollow Spheroids
As we have shown, mitonucleons within single cells result in floating spheroids.
Spheroids have been of interest as an alternative method for culturing mammalian cells for
almost as long as scientists have known how to keep human cells alive in monolayers. Most
papers describing cell spheroids are concerned with solid spheroids such as are shown in fig.
9. It is thought that cells interacting with each other in such a three-dimensional structure
may more closely mimic the condition of cells in vivo
. Whereas the terminology for
monolayers is relatively simple, the terminology used for spheroids has led to at least 25
different terms for these three dimensional structures , many deriving from tissue origins,
with a nod to the methodology used to foster their formation (Weiswald et al.; 2015). Most
cell lines such as Ishikawa can be grown as monolayers or as spheroids by manipulating growth
conditions. Solid spheroids were elicited in cultures of Ishikawa cells plated with medium
containing no serum. As Achilli et al. (2012) recently described, there are at least 7 other
culturing conditions that will result in the formation of spheroids by lessening contact
between cells and surfaces such as the plastic surface of petri dishes. Furthermore, and
relevant to our discussion, there is at least one paper showing that ovarian cancer cells in a
monolayer can give rise to multicellular solid spheroids spontaneously and that these are
released into the medium following a period of being tethered to the monolayer that
produced them (Pease et al. 2012).
A review article in 2008 discusses the biological features unique to spheroids that might make
them an important player in dissemination of ovarian epithelial cancer. (Shields et al., 2008)
One of the most fascinating differences demonstrated between solid spheroids and
monolayers of epithelial ovarian cells appears to be that many fewer cells disaggregated from
a spheroid will elicit cancer in a test animal than cells harvested from a monolayer (Odunsi et
al., 2014). As few as 2000 cells from a dissociated spheroid of epithelial ovarian cancer cells
will, when injected into immune deficient mice, elicit the formation of a tumor. Even a bolus
of parent cells from monolayers five times larger than that was not found to be consistently
tumorigenic. This observation, and the results from a variety of assays for stem cells, led the
researchers to conclude that solid spheroids are enriched in a population of stem cells that
are highly tumorigenic.
But there are differences among spheroids. Most of the results listed in the NIH database for
cellular spheroids, close to 5000, are about solid spheroids. Fewer than 10% of that number
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
discuss hollow spheroids most of which are found in malignant ascites fluid. Furthermore, and
important for the results in this paper, most of the time the term hollow spheroid has been used
to describe multicellular hollow spheroids in which a single layer of 50-200 epithelial cells
surrounds the inner lumen. In this paper we have demonstrated a third kind of spheroid which
is both unicellular and polyploid.
The earliest paper on multicellular hollow spheroids was published in 1979 by a Japanese
research team (Nishida et al.) who called hollow spheroids “mirror balls.” In 1987, Allen’s group
(Allen et al.) from Roswell Park published a paper with pictures of hollow spheroids in ascites
fluid. They reported on multicellular aggregates found in the peritoneal fluid of patients with
ovarian cancer, describing clusters ranging “from compact to loosely adherent groups of cells to
spheroids with a central lumen surrounded by a cell monolayer” . Allen alludes to observations
of acellular spheres of light density but concludes that these result from a loss of surrounding
monolayer cells during processing of the fluid. Maybe, or maybe he was seeing structures like
those pictured in fig. 8, which appear acellular if the microscope is not focused on the nuclei (fig.
8b), a phenomenon that would not have been anticipated.
More than a couple of decades passed before many more studies were done. Ishiwata and
colleagues (1997) have studied ovarian and endometrial adenocarcinoma cells in vivo
and in
vitro
. Both cell types formed “hollow cell ball structures” (their term for multicellular hollow
spheroids) but the researchers concluded that these arose from solid cell ball structures,
hollowed out by necrosis of the internal cells. What they saw may have arisen in that manner.
The unicellular hollow spheroids that are the subject of this paper definitely do not start out as
solid spheroids, although they are sometimes seen budding out from solid spheroids.
Much of the work of Kato and colleagues (2011) involving hollow spheroids harvested from
patients with clear cell carcinoma, attempted to correlate the presence of spheroids with the
presence of spherule-like hyaluronan rich stroma in tumor tissue. The researchers also
collected spheroids from the abdominal cavity of nuce mice and demonstrated that hollow
spheroids survive in vitro and give rise to daughter hollow spheroids. The evidence in Ishikawa
cultures does not contradict those findings but most clearly shows that solid spheroids of
Ishikawa endometrial epithelia can give rise to unicellular hollow spheroids.
Interestingly enough hollow spheroids, embedded in normal tissue, have also been found
beyond the invasive margin of tumors of the colon and have been identified as an
independent risk factor for metastases and shorter survival. (Tamura et al. 2011) It is not
obvious how these hollow spheroids develop within tissue but the implication is that they
arise from invading cells that are vectors of malignancy. That observation suggests that there
may be significant relevance from the point of view of metastasizing solid tumors, that
unicellular hollow spheroids can “bud” out from solid spheroids in vitro.
The unicellular hollow spheroids being described in this paper arise from single cells that have
become multinucleated and enriched in structures containing biotin. When such a cell is
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
detaching from the colony in which it was created, mitonucleon(s) can be detected in the center
of the cell. As recently shown, mitonucleons retain gases (Fleming, 2018). An increasing gas
vacuole results in the creation of a structure that looks like the three dimensional realization of
“signet cells,” with one or more polyploid nuclei and cytosol compressed in the space between
two membranes: the outer membrane of the mitonucleon ( vacuoles having fused as
mitonucleons became distended) and the cell membrane. Gases build up pressing cell contents
against the cell membrane. The resulting highly unusual cell structure floats around until
conditions are appropriate to once again become attached. Since endometrial cells that
detach from the uterus and move out into the peritoneal fluid seem to be the most likely
vectors for endometriosis, the mechanisms that such cells use to switch from being attached
to being unattached and mobile are of interest. Furthermore the observation of hollow
spheroids (Allen et al.) in malignant ascites fluid may be relevant to the metastases of
ovarian cancer.
Materials and Methods
It is relevant that the unicellular hollow spheroids described in this report would probably be destroyed
by many of the methods used to prepare cell specimens for microscopy. The handling of medium
containing unicellular hollow spheroids was kept to a minimum. All of the photomicrographs in this
paper are of cultured endometrial epithelia called Ishikawa developed by Nishida and colleagues in
1985. His laboratory established that the line contained receptors for both of the female sex steroids,
estradiol and progesterone. The cells were obtained from Erlio Gurpide’s laboratory at Mt. Sinai, New
York. As described in the first paper on the topic of dome formation in 1995 (Fleming), the cells were
cultured in phenol red-free Minimum Essential Medium (MEM) supplemented with 2mM glutamine, 100
U/ml penicillin, 0.1 mg/ml and .25 mg amphotericin B (GIBCO, Grand Island, NY).
Multinucleated cells were found in cultures approaching confluence. Cells seeded at an
approximate density of 5 x 105cells/cm2 in MEM containing 5% calf serum (CS) at 37 degrees C and 5%
CO2 were grown for at least one week. Cells were viewed using an Olympus inverted stage
microscope at powers of 100X (fig1a), 200X (fig.1b) and 400X (all other figures).
The structures shown in figs. 2b and 6 through 11 were photographed in living cultures.
As first described in a paper published in 1998 (Fleming et al.) endogenous biotin in the structures
shown in figs. 1, 2a, 3, 4, and 5 was detected by an assay using extra-avidin conjugated horseradish
peroxidase. Dishes of cells were fixed by adding 4% paraformaldehyde in phosphate buffered saline
(PBS) to culture dishes at various times after they had been stimulated to form domes. After 10
min, the cells were washed gently four times with 5-10 ml PBS. A solution of 1% Triton X-100 was
added to the cells to permeabilize the membrane. Again after 5 min, the culture was washed with
successive changes of PBS. After washing, cells were exposed to a 1:200 dilution of
Extravidin-conjugated horseradish peroxidase (HRP) (Sigma) for 30 min. After further washing with
PBS , a solution of 3-amino-9-ethylcarbazole (20 mg of AEC in 2.5 ml of dimethylformamide, diluted
with 47.5 ml of 50 mM potassium acetate adjusted to pH 5.0) was added to the cells together with
.25% H2O2. This solution was incubated at 37"C for 45 min to allow color to develop. The AEC
solution was removed, and the cultures were examined anda stored in the presence of PBS at 4"C.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Additionally cells were stained with a solution of hematoxylin and eosin. All of the operations were
carried out with care not to expose the monolayer or material above the monolayer to air.
If avidin linked to peroxidase is not added to the cultures, there is no reaction. If avidin without
peroxidase is added first to the cultures, followed by avidin-linked to peroxidase, staining is not
observed. Staining does not occur if avidin-HRP is not added to the cultures prior to AEC indicating
that an endogenous peroxidase is not responsible for the staining. To ensure that avidin was
reacting with biotin, we stained domes using streptavidin linked to horseradish peroxidase as well as
primary antibody to biotin and secondary antibody-linked to horseradish peroxidase. Staining
occurred under all circumstances, indicating that avidin does indeed react with biotin that is
endogenously present in cells in significant amounts.
References
Achilli, T.-M., Meyer, J., & Morgan, J. R. 2012. Advances in the formation, use and understanding of
multi-cellular spheroids. Expert Opinion on Biological Therapy
, 12
(10), 1347–1360.
http://doi.org/10.1517/14712598.2012.707181
Allen HJ, Porter C, Gamarra M, Piver MS, Johnson EAZ. 1987. Isolation and morphologic
characterization of human ovarian carcinoma cell clusters present in Effusions. Expl Cell Biol
1987;55:194-208.
Armstrong, EM., More, IA., McSeveney, D., & Carty, M. 1973. The giant mitochondrion-endoplasmic
reticulum unit of the human endometrial glandular cell. Journal of Anatomy
, 116
(Pt 3), 375–383.
Boxma B1, de Graaf RM, van der Staay GW, van Alen TA, Ricard G, Gabaldón T, van Hoek AH, Moon-van
der Staay SY, Koopman WJ, van Hellemond JJ, Tielens AG, Friedrich T, Veenhuis M, Huynen MA,
Hackstein JH. 2005. An anaerobic mitochondrion that produces hydrogen. Nature 434(7029):74-9.
Embley, T. M., van der Giezen, M., Horner, D. S., Dyal, P. L., & Foster, P. 2003. Mitochondria and
hydrogenosomes are two forms of the same fundamental organelle. Philosophical Transactions of the
Royal Society B: Biological Sciences
, 358
(1429), 191–202. http://doi.org/10.1098/rstb.2002.1190
Endeward V, Gros G. 2013. Low carbon dioxide permeability of the apical epithelial membrane of
guinea-pig colon. Front Physiol. 4:382.
Endeward V, Al-Samir S, Itel F, Gros G. 2014 How does carbon dioxide permeate cell membranes? A
discussion of concepts, results and methods. J Physiol. 567(Pt 1):253-65. 384 Epub 2005 Jun 2.
Erenpreisa JA1, Cragg MS, Fringes B, Sharakhov I, Illidge TM. 2000 Release of mitotic descendants by
giant cells from irradiated Burkitt's lymphoma cell line. Cell Biol Int. 2000;24(9):635-48.
Erenpreisa J1, Salmina K, Huna A, Kosmacek EA, Cragg MS, Ianzini F, Anisimov AP. 2011.
Polyploid tumour cells elicit paradiploid progeny through depolyploidizing divisions and regulated
autophagic degradation. Cell Biol Int. 35(7):687-95. doi: 10.1042/CBI20100762.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Fleming H. 1995. Differentiation in human endometrial cells in monolayer culture: dependence on a
factor in fetal bovine serum. Journal of Celluar Biochemistry. 57(2): 262- 391 270.
Fleming H, Condon R, Peterson G, Guck I, Prescott E, Chatfield K, Duff M. 1998. Role of 398
biotin-containing membranes and nuclear distribution in differentiating human endometrial 399 cells.
Journal of Cellular Biochemistry. 71(3): 400-415.
Fleming H. 1999. Structure and function of cultured endometrial epithelial cells. Semin Reprod
Endocrinol. 17(1):93-106
Fleming H. 2014. Unusual characteristics of opaque Ishikawa endometrial cells include the envelopment
of chromosomes with material containing endogenous biotin in the latter stages of cytokinesis DOI
10.7287/peerj.preprints.772v1
Fleming H. 2016a. Mitonucleons formed during differentiation of Ishikawa endometrial cells generate
vacuoles that elevate monolayer syncytia: Differentiation of Ishikawa domes, Part 1. PeerJ PrePrints
4:e1728v1
Fleming H. 2016b. Pyknotic chromatin in mitonucleons elevating in syncytia undergo 408 karyorhhexis
and karyolysis before coalescing into an irregular chromatin mass: Differentiation of Ishikawa
Endometrial Cells, Parts 2, submitted to PeerJ as a preprint.
Fleming H. 2016c. Amitosis generates differentiated cell nuclei from chomatin mass formed after
monolayer nuclei are aggregated, become pyknotic, and are fragmented: Differentiation of Epithelial
Domes, Part 3. submitted to PeerJ as a preprint.
Fleming H. 2018. Mitochondrial/Nuclear Superstructures Drive Morphological Changes in Endometrial
Epithelia by Pressure Exerted when Gas vacuoles Form and Coalesce Within Superstructures. In press
Gamachi A, Kashima K, Daa T, Nakatani Y, Tsujimoto M, Yokoyama S. 2003 Aberrant intranuclear
localization of biotin, biotin-binding enzymes, and beta-catenin in pregnancy-related endometrium and
morule-associated neoplastic lesions. Mod Pathol. 16(11):1124-31.31.
Itel F, Al-Samir S, Öberg F, Chami M, Kumar M, Supuran CT, Deen PM, Meier W, Hedfalk K, Gros G,
Endeward V. 2012. CO2 permeability of cell membranes is regulated by membrane cholesterol and
protein gas channels. FASEB J. 26(12):5182-91 Sep 10.
Ishiwata I, Kiguchi K, Ishiwata C, Soma M, Nakaguchi T, Ono I, Tachibana T, Hashimoto H, Ishikawa H,
Nozawa S. 1997. Histogenesis of hollow cell ball structure of ovarian and endometrial adenocarcinoma
cells in vivo and in vitro. Hum Cell. 10(3):209-16.
Kato N, Narutomi K, Fukase M, Motoyama T. Hollow spheroids in ascites of ovarian clear cell
carcinoma: how are they formed and how do they behave? Cytopathology. 2012 Apr;23(2):120-5. doi:
10.1111/j.1365-2303.2011.00847.x. Epub 2011 Feb 8. PubMed PMID: 21306451.
Kornberg TB. & Sougata R. 2014. Cytonemes as specialized signaling filopodia. Development 141
729-736 doi: 101242/dev.086223
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
Liao J1, Qian F2, Tchabo N3, Mhawech-Fauceglia P4, Beck A5, Qian Z5, Wang X6, Huss WJ7, Lele SB8,
Morrison CD4, Odunsi K9.PLoS One. 2014. Ovarian cancer spheroid cells with stem cell-like properties
contribute to tumor generation, metastasis and chemotherapy resistance through hypoxia-resistant
metabolism.9(1):e84941. doi: 10.1371/journal.pone.0084941. eCollection 2014.
Nakhoul NL, Davis BA, Romero MF & Boron WF. 1998. Effect of expressing the water channel
aquaporin-1 on the CO2 permeability of Xenopus oocytes. Am J Physiol 274, C543-C548.
Nishida M, Kasahara K, Kaneko M, Iwasaki H. 1985. Establishment of a new human endometrial
adenocarcinoma cell line, Ishikawa cells, containing estrogen and progesterone receptors. Acta Obstet
Gynaec Japonica (In Japanese)37:1103-1111
Nishida M, Kuramoto H, Tatsumi H, Miura T, Araj M. 1979. Cytological findings of ascites in patients
with ovarian clear cell carcinoma. Diagnostic significance and genesis of mirror ball pattern. Nippon
Rinsho Saibo Gakkai Zasshi a8: 280-294 (In Japanese with English Abstr)
Painter, T. S. 1959. The Elimination of DNA from Soma Cells. Proceedings of the National Academy of
Sciences of the United States of America
, 45
(6), 897–902.
Pease, J. C., Brewer, M., & Tirnauer, J. S. 2012. Spontaneous spheroid budding from monolayers: a
potential contribution to ovarian cancer dissemination. Biology Open
, 1
(7), 622–628.
http://doi.org/10.1242/bio.2012653
Shield K, Ackland ML, Ahmed N, Rice GE. Multicellular spheroids in ovarian cancer metastases: Biology
and pathology. Gynecol Oncol. 2009. 113(1):143-8. doi: 10.1016/j.ygyno.2008.11.032. Epub 2009 Jan
10. Review. PubMed PMID: 19135710.
Sundaram M1, Guernsey DL, Rajaraman MM, Rajaraman R. 2004. Neosis: a novel type of cell division in
cancer.10.15252/embj.201488104. Epub 2014 Aug 8. Cancer Biol Ther.;3(2):207-18. Epub 2004 Feb 1.
Tamura, K., Yokoyama, S., Ieda, J., Takifuji, K., Hotta, T., Matsuda, K., Yamaue, H. 2011. Hollow
spheroids beyond the invasive margin indicate the malignant potential of colorectal cancer. BMJ Open
,
1
(1), e000179. http://doi.org/10.1136/bmjopen-2011-000179
Waisbren SJ, Geibel JP, Modlin IM, Boron WF. 1994. Unusual permeability properties of gastric gland
cells. Nature. 368(6469):332-5.
Weiswald, L.-B., Bellet, D., & Dangles-Marie, V. 2015. Spherical Cancer Models in Tumor Biology.
Neoplasia (New York, N.Y.)
, 17
(1), 1–15. http://doi.org/10.1016/j.neo.2014.12.004
Zybina EV, Zybina TG. 2008, Modifications of nuclear envelope during differentiation and
depolyploidization of rat trophoblast cells. Micron.39(5):593-606. Epub 2007
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26793v1 | CC BY 4.0 Open Access | rec: 29 Mar 2018, publ: 29 Mar 2018
