Thyroid Organoids Made from Stem Cells Treat Thyroid-Deficient Mice

Darrell Kotton and his research team from Beth Deaconess Medical Center, in collaboration with researchers from the Boston University School of Medicine have devised a workable protocol for differentiating Human pluripotent stem cells into functional thyroid gland cells.

Every year, many people are diagnosed with an underactive thyroid and many others lose their thyroid as a result of thyroid cancer. Designing treatments that can help replace lost thyroid tissue would certainly be a welcome thing for these patients.

By working with mouse embryonic stem cells, Kotton and his colleagues showed that two growth factors, BMP4 and FGF2, and induce foregut endodermal cells to differentiate into thyroid cells. This simple signaling pathway not only efficiently generates thyroid tissue from endoderm, but this pathway turns out to be commonly used in species as diverse as frogs, mice and humans.

The BMP4/FGF2-treated foregut cells differentiated into small thyroid organics that Kotton and his team were able to transplant into thyroid-deficient mice. These transplantations restored normal thyroid function to these mice.

fx1 (6)

While mice cells are a fine model system for human diseases, they are not exactly the same. Can this procedure work with human cells? To answer that question, Kotton and his coworkers used human induced pluripotent stem cells (iPSCs) and subjected them to the same BMP4/FGF2 protocol after they had first differentiated the cells into endoderm. In addition, Kotton and his team made thyroid cells from iPSCs derived from cells taken from patients with a specific type of hypothyroidism (interactive thyroid). These patients lack a gene called NKX2-1, and suffer from congenital hypothyroidism.

The thyroid is responsible for your basal metabolic rate. Hypothyroidism or an interactive thyroid can cause patients to gain weight, feel tired constantly, have trouble concentrating, and have a slow heart rate. Hypothyroidism is usually treated with synthetic thyroid hormones that are taken orally. However, restoring a patient’s own thyroid tissue or even replacing defective thyroid tissue with repaired thyroid tissue would be a huge boon to thyroid patients.

This work has discovered the regulatory mechanisms that drive the establishment of the thyroid. It also provides a significant step toward cell-based regenerative therapy for hypothyroidism and the replacement of the thyroid after thyroid cancer treatments.

These results were published in the journal Cell Stem Cell, October 2015 DOI:10.1016/j.stem.2015.09.004.

Infertility Treatment with Stem Cells is Unlikely

Because several laboratories have managed to differentiate embryonic stem cells into cells that look very much like human eggs and sperm, many have predicted that infertility will be treated with stem cell treatments (see Volarevic V, et al., Biomed Res Int. 2014;2014:507234). However, new work from the University of Gothenburg and Karolinska Institute has cast doubt on this hope.

At about 24 days of life, large, spherical sex cells are recognizable among the endodermal cells of the umbilical vesicle close to the allantois. These cells are the primordial germ cells and they are the progenitor cells of the sperm in men and eggs in women. As the embryo folds during about the late 4th week, the dorsal portion of the umbilical vesicle is incorporated into the embryo. This incorporation of the umbilical vesicle occur concurrently with the migration of these primordial germ cells along the dorsal mesentery of the hindgut to the gonadal ridges. During the 6th week of life, the primordial germ cells enter the underlying mesenchyme and are incorporated into the gonadal cords. Primordial germ cell migration is mainly regulated by three genes: Stella, Fragilis, and BMP-4.

PGC migration

These primordial germ cells divide as they migrate, and by five months of gestation, embryonic ovaries contain about six to seven million oogonia. Most of these oogonia experience cell death before birth, but the remaining oogonia begin meiosis toward the end of gestation. At this time, the oogonia are called primary oocytes. Meiosis is arrested in prophase of the first meiotic division, and this is the same stage at which spermatogenesis in the male is blocked. Primary oocytes decrease in number throughout a woman’s life. The ovaries of a newborn girl contain about two million primary oocytes and these are all the gametes she will ever have. Each primary oocyte is contained within a hollow ball of cells called the ovarian follicle. By the time a woman reaches puberty, that number of primary oocytes has been reduced to 400,000. Only about 400 of these cells will ovulate during a woman’s reproductive years. The rest will die by means of programmed cell death. Once all the primary oocytes are gone, ovulation stops and the woman undergoes menopause.

Kui Liu from the University of Gothenburg said: “Ever since 2004, the studies on stem cell research and infertility have been surrounded by hype. There has been a great amount of media interest in this, and the message has been that the treatment of infertility with stem cells is about to happen. However, many researchers, including my research group, have tried to replicate these studies and not succeeded. This creates uncertainty about whether it is all possible to create new eggs with the help of stem cells.”

In collaboration with Outi-Hovatta’s laboratory at the Karolinska Institute and Jan-Åke Gustafsson’ research team at the university of Houston in the US, Lui’s research team carried out experiments on mice that failed to demonstrate that functional gametes could be formed from pluripotent stem cells. Essentially, the only gametes that could that the female mice had were the ones they were born with.

In Liu’s opinion, fertility clinics should place their attention on using the eggs that women still have in order to treat infertility.

Foregut Stem Cells

Scientists from Cambridge University have designed a new protocol that will convert pluripotent stem cells into primitive gut stem cells that have the capacity to differentiate into liver, pancreas, or some other gastrointestinal structure.

Nicholas Hannan and his colleagues at the University of Cambridge Welcome Trust MRC Stem Cell Institute have developed a technique that allows researchers to grow a pure, self-renewing population of stem cells that are specific to the human foregut, which is the upper section of the human digestive system. These types of stem cells are known as “foregut stem cells” and they can be used to make liver, pancreas, stomach, esophagus, or even parts of the small intestine. Making these types of gastrointestinal tissues can provide material for research into gastrointestinal abnormalities, but might also serve as a source of material to treat type 1 diabetes, liver disease, esophageal and stomach cancer, and other types of severe gastrointestinal diseases.

“We have developed a cell culture system which allows us to specifically isolate foregut stem cells in the lab,” said Hannan. “These cells have huge implications for regenerative medicine, because they are the precursors to the thyroid upper airways, lungs, liver, pancreas, stomach, and biliary systems.”

Hannan did this work in the laboratory of Ludovic Vallier, and they think that their technique will provide the means to analyze the precise embryonic development of the foregut in greater detail. “We now have a platform from which we can study the early patterning events that occur during human development to produce intestines, liver, lungs, and pancreas,” said Hannan.

To make foregut stem cells, Hannan begins with a pluripotent stem cell line; either an embryonic stem cell line or an induced pluripotent stem cell line. Then he differentiated them into definitive endoderm by treating them with CDM-PVA and activin-A (100 ng/ml), BMP4 (10 ng/ml), bFGF (20 ng/ml), and LY294002 (10 mM) for 3 days. Once they differentiated into endoderm, the endodermal cells were grown in RPMI+B27 medium with activin-A (50 ng/ml) for 3-4 days in order to generate foregut stem cells.

(A) GFP-expressing hPSCs were differentiated into hFSCs. (B) Single GFP-positive hFSCs were seeded onto a layer of non-GFP hFSCs and then expanded for five passages. The resulting population was then split into culture conditions inductive for liver or pancreatic differentiation. (C and D) GFP-hFSCs differentiated for 25 days were found to respectively generate cells expressing liver markers (ALB, LDL-uptake) and pancreatic markers (PDX1, C-peptide) from both hESC-derived (C) and hIPSC-derived (D) hFSCs.
(A) GFP-expressing hPSCs were differentiated into hFSCs.  (B) Single GFP-positive hFSCs were seeded onto a layer of non-GFP hFSCs and then expanded for five passages. The resulting population was then split into culture conditions inductive for liver or pancreatic differentiation.  (C and D) GFP-hFSCs differentiated for 25 days were found to respectively generate cells expressing liver markers (ALB, LDL-uptake) and pancreatic markers (PDX1, C-peptide) from both hESC-derived (C) and hIPSC-derived (D) hFSCs.

These foregut stem cells (FSCs) can self-renew, and can also differentiate into any part of the foregut. Thus, FSCs can grow robustly in culture, and they can also differentiate into foregut derivatives. However, these cells also do not form tumors. When injected into mice, they failed to form tumors.

(A) Large cystic hFSC outgrowth under the kidney capsule of a NOD-SCID mouse. (B) Cryosection of a hFSC outgrowth showing large cystic structures lined with epithelial cells. (C) Immunocytochemistry showing foregut outgrowths expressing EpCAM, PDX1, AFP, and NKX2.1. Scale bars, 100 μm or 50 μm as shown. See also Figure S4.
(A) Large cystic hFSC outgrowth under the kidney capsule of a NOD-SCID mouse.  (B) Cryosection of a hFSC outgrowth showing large cystic structures lined with epithelial cells.  (C) Immunocytochemistry showing foregut outgrowths expressing EpCAM, PDX1, AFP, and NKX2.1.  Scale bars, 100 μm or 50 μm as shown. See also Figure S4.

What are the advantages to FSCs as opposed to making pancreatic cells or liver cells from pluripotent stem cells? These types of experiments always create cultures that are impure. Such cultures are difficult to use because not all the cells have the same growth requirements and they would be dangerous for therapeutic purposes because they might contain undifferentiated cells that might grow uncontrollably and cause a tumor. Therefore, FSCs provide a better starting point to make pure cultures of pancreatic tissues, liver tissues, stomach tissues and so on.

Ludovic Vallier, the senior author of this paper said this of his FSCs, “What we have now is a better starting point – a sustainable platform for producing liver and pancreatic cells. It will improve the quality of the cells that we produce and it will allow us to produce the large number of uncontaminated cells we need for the clinical applications of stem cell therapy.”

Vallier’s groups is presently examining the mechanisms that govern the differentiation of FSCs into specific gastrointestinal cell types in order to improve the production of these cells for regenerative medicine.

Microparticles and Local Control of Stem Cells

Using stem cells to grow three-dimensional structures, such as organs or damaged body parts, requires that scientists have the ability to control the growth and behavior of those cells. Also, adapting such a technology to an off-the-shelf kind of process so that it does not cost an arm and a leg is also important.

A research project by scientists from Atlanta, Georgia has used gelatin-based microparticles to deliver growth factors to specific areas of aggregates of stem cells that are differentiating. This localized delivery of growth factors provides spatial control of cell differentiation, which enables the creation of complex, three-dimensional tissues. The local delivery of growth factors also decreases the amount of growth factor used and, consequently, the cost of the procedure.

This particular microparticle technique was used on mouse embryonic stem cells and it proved to provide better control over the kinetics of cell differentiation since it delivered that promote cell differentiation or inhibit it.

Todd McDevitt, associate professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University, said, “By trapping these growth factors within microparticle materials first, we are concentrating the signal they provide to the stem cells. We can then put the microparticle materials physically inside the multicellular aggregate system that we use for differentiation for the stem cells. We have good evidence that this technique can work, and that we can use it to provide advantages in several areas.”

The differentiation of stem cells is largely controlled by external cues, including protein growth factors that direct cell proliferation, and differentiation that are available in the three-dimensional environment in which the cells live. In most experiments, stem cells are grown in liquid culture and growth factor is equally accessible to the growth factors. This makes the cultures quite homogeneous. But delivering the growth factors via microparticles gives better control of the spatial and temporal presentation of these growth factors to the stem cells. This gives scientists the means to make heterogeneous structures from stem cell cultures.

When embryonic stem cells grow in culture, they tend to clump together. When the growth medium is withdrawn or if growth factors that induce differentiation are added, the cells form an “embryoid body” that is stuff with cells differentiating into all kinds of cell types. When McDevitt and his co-workers added microparticles with the growth factors BMP4 (bone morphogen protein 4) or Noggin (which inhibits BMP4 signaling), they centrifuged the cells and found that the microparticles found their way into the interior of the embryoid bodies.

When they examined the embryoid bodies, with confocal microscopy they found that BMP4 directed the cells to make mesodermal and endodermal derived cell types. However, because the microparticles were in direct contact with the cells, they needed 12 times less growth factor than was required by solution-based techniques.

“One of the major , in a practical sense, is that we are using much less growth factor,” said McDevitt. “From a bioprocessing standpoint, a lot of the cost involved in making stem cell products is related to the cost of the molecules that must be added to make the stem cells differentiate.”

Beyond more focuses signaling, the microparticles also provided localized control that was not available through other techniques. It allowed researchers to create spatial differences in the aggregates and this is an important possible first step toward forming more complex structures with different tissue types such as vascularization and stromal cells.

“To build tissues, we need to be able to take stem cells and use them to make many cell types which are grouped together in particular spatial patterns,” explained Andres M. Bratt-Leal, the paper’s first author and a former graduate student in McDevitt’s lab. “This spatial patterning is what gives the ability to perform higher order functions.”

Once the stem cell aggregates were made and treated with growth factor-endowed microparticles, McDevitt and his colleagues saw spheres of cells with differentiating cells.

“We can see the microparticles had effects on one population that were different from the population that didn’t have the particles,” said McDevitt. “This may allow us to emulate aspects of how development occurs. We can ask questions about how tissues are naturally patterned. With this material incorporation we have the ability to better control the environment in which these cells develop.”

The microparticles could provide better control over the kinetics of cell differentiation; slowing it down with molecules that antagonize differentiation or speed up with other molecules that promote stem cell differentiation.

Despite the fact that McDevitt and his colleagues used mouse embryonic stem cells in this paper, he and his co-workers are already testing this technology on human embryonic stem cells, and the results have been comparable.

“Our findings will provide a significant new tool for tissue engineering, bioprocessing of stem cells and for better studying early development processes such as axis formation in embryos,” said Bratt-Leal. “During development, particular tissues are formed by gradients of signaling molecules. We can now better mimic these signal gradients using our system.”

Developmental Regression: Making Placental Cells from Embryonic Stem Cells

A research group from Copenhagen, Denmark has discovered a way to make placental cells from embryonic stem cells. In order to do this, the embryonic stem cells must be developmentally regressed so that they can become wither placenta-making cells rather than inner cell mass cells.

This study is significant for two reasons. First of all, it was thought to be impossible to make placental cells from embryonic stem cells because embryonic stem cells (ESCs) are derived from the inner cell mass cells of 4-5-day old human blastocysts. These early embryos begin as single-celled embryos that divide to form 12-16-cell embryos that undergo compaction. At this time, the cells on the outside become trophoblast cells, which will form the trophectoderm and form the placenta and the cells on the inside will form the inner cell mass, which will form the embryo proper and a few extraembryonic structures. Since ESCs are derived from inner cell mass cells that have been isolated and successfully cultured, they have already committed to a cell fate that is not placental. Therefore, to differentiate ESCs into placental cells would require that ESCs developmentally regress, which is very difficult to do in culture.

Secondly, if this could be achieved, several placental abnormalities could be more easily investigated, For example, pre-eclampsia is a very serious prenatal condition that is potentially fatal to the mother, and is linked to abnormalities of the placenta. Studying a condition such as pre-eclampsia in a culture system would definitely be a boon to gynecological research.

Because human ESCs can express genes that are characteristic of trophoblast cells if they are treated with a growth factor called Bone Morphogen Protein 4 (BMP4), it seems possible to make placental cells from them (see Xu R.H., Chen X., Li D.S., Li R., Addicks G.C., Glennon C., Zwaka T.P., Thomson J.A. BMP4 initiates human embryonic stem cell differentiation to trophoblast. Nat. Biotechnol. 2002;20:1261–1264, and Xu RH. Methods Mol Med. 2006;121:189-202). However, a study by Andreia S. Bernardo and others from the laboratory of Roger Pedersen at the Cambridge Stem Cell Institute strongly suggested that BMP4 treatment, even in the absence of FGF signaling (another growth factor that has to be absent for BMP4 to induce trophoblast-like gene expression from ESCs) the particular genes induced by BMP4 are not exclusive to trophoblast cells and more closely resemble mesodermal gene profiles (see AS Bernardo, et al., Cell Stem Cell. 2011 Aug 5;9(2):144-55).

Into the fray of this debate comes a paper by stem cells scientists at the Danish Stem Cell Center at the University of Copenhagen that shows that it is possible to rewind the developmental state of ESCs.

In this paper, Josh Brickman and his team discovered that if they maintained mouse ESCs under specific conditions, they could cause the cells to regress into very early pre-blastocyst embryonic cells that can form trophoblast cells or ICM cells.

“It was a very exciting moment when we tested the theory, said Brinkman. “We found that not only can we make adult cells but also placenta, in fact we got precursors of placenta, yolk sac as well as embryo from just one cell.”

“This new discovery is crucial for the basic understanding of the nature of embryonic stem cells and could provide a way to model the development of the organism as a whole, rather than just the embryonic portion,” said Sophie Morgani, graduate student and first author of this paper. “In this way we may gain greater insight into conditions where extraembryonic development is impaired, as in the case of miscarriages.”

To de-differentiate the ESCs, Brinkman and his colleagues grew them in a solution called “2i.”  This 2i culture medium contained inhibitors of MEK and GSK3.  MEK is a protein kinase that is a central participant in the “MAP kinase signaling pathway, which is a signaling pathway that is central to cell growth and survival.  This particular signaling pathway is the target of the anthrax toxin, which illustrates its importance,  GSK3 stands for “glycogen synthase kinase 3,” which is a signaling protein in the Wnt pathway.

When the mouse ESCs were grown in 2i medium they expressed genes normally found only in pre-blastocyst embryos (Hex, for example).  Therefore, the 2i medium directs mouse ESCs to de-differentiate.  When ESCs grown in 2i were implanted into mouse embryos, they divided and differentiated into cells that were found in placental and embryonic fates.  This strongly argues that the ESCs grown in 2i became pre-blastocyst embryonic cells.  When the ESCs grown in 2i were also grown with LIF, which stands for “leukemia inhibitory factor” (LIF is a protein required for the maintenance of mouse ESCs in culture), the 2i cells were maintained in culture and grew while maintaining their pre-blastocyst status.  These cells differentiated into placental cells, embryonic or fetal cells.  Essentially, the 2i-cultured cells when from being pluripotent to being “totipotent,” or able to form ALL cell types in the embryo, fetus, or the adult.

ESC de-differentiation in totipotence

“In our study we have been able to see the full picture unifying LIF’s functions: what LIF really does, is to support the very early embryo state, where the cells can make both embryonic cells and placenta. This fits with LIFs’ role in supporting implantation,” said Brinkman.

This study definitively shows that ESCs are NOT embryos.  ESCs can regress in their development but embryos develop forward, becoming more committed as they develop and more restricted in the cell fates they can form.  This should effectively put the nail in the coffin of Lee Silver’s argument against Robert P. George that embryonic stem cells are embryos.  They are definitely and unequivocally, since embryos do NOT develop in reverse, but ESCs can and do.

Robert P. George argues that early human embryos, like the kind used to make ESCs are very young  members of the human race and deserve, at the minimum, the right not to be harmed.  Silver counters that George’s argument is inconsistent because George would not extend the same right to an ESC cell line, which is the same as an embryo.  His reasoning is that mouse ESCs can be transplanted into other mouse embryos that have four copies of each chromosome.  The messed up mouse embryo will make the placenta and the ESCs will make the inner cell mass and the mouse will develop and even come to term.  This is called tetraploid rescue, and Silver thinks that this procedure is a minor manipulation, but that it shows that ESCs are functionally the same as embryos.

I find Silver’s argument wanting on just about all fronts.  This is not a minor manipulation.  The tetraploid embryo is bound for certain death, but the implanted ESCs use the developmental context of the tetraploid embryo to find their place in it and make the inner cell mass.  The ESCs do not do it all on their own, but instead work with the tetraploid embryo in a complex developmental give-and-take to make an embryo with the placenta from one animal and the embryo proper from another.

Thus Silver’s first argument does not demonstrate what he says it does.  All it demonstrates is that ESCs can contribute to an embryo, which is something we already knew and expected.  This new data completes blows Silver’s assertion out of the water, since ESCs can take developmental steps backward and embryos by their very nature and programming, do not.  Thus these two entities are distinct entities and are not identical.  The early embryo is a very young human person, full stop.  We should stop dismembering them in laboratories just to stem our scientific curiosity.