|
Theory of TransdifferentiationTransdifferentiation means the conversion of cells from one differentiated cell type to another. Some authors have considered that it does not occur at all in nature. There is however good evidence that it does occur in some cases, particularly in situations where missing parts regenerate in animals (Okada, 1991). One well known example is the Wolffian regeneration of the lens of various species of urodele amphibia (newts and salamanders). Here, after removal of the lens of the eye, a new lens regenerates from the dorsal iris (Tsonis et al. 2004). This is a transdifferentiation because the iris cells are pigmented epithelial cells very similar to those of the pigmented retina. The lens on the other hand is composed of modified keratinocytes containing high concentrations of crystallin proteins that give the lens its characteristic transparency. Another well established instance of transdifferentiation is the regeneration of striated muscle in the jellyfish (Schmid 1992). If the definition is relaxed a little to include transformation between undifferentiated cells committed to form specific tissue types or body parts, then many more examples appear. They have been given various different names in the past but all refer to the same basic type of process. Serial heteromorphosis, transdetermination, metaplasia.Serial heteromorphosis is found in the regeneration of appendages in crustacea and insects. For example in the shrimp Palinurus, if the claw is amputated it may regenerate not as a claw but as an antenna. In several types of insect, an amputated antenna may be replaced by parts of a leg (Villee, 1942; Needham 1952). Transdetermination is the name given to
transformations between different imaginal discs in Drosophila.
Imaginal discs are formed as small clusters of cells in the Drosophila
embryo. During larval life they grow considerably in size but do not
differentiate. During pupal life they differentiate into the specific
segmental appendages or parts of segments, for example antennae, wings
and legs (Cohen 1993). Imaginal discs can survive in the undifferentiated
state indefinitely if they are transplanted into the abdomen of an adult.
Here they are nourished by the haemolymph of the host and can grow indefinitely
without differentiating. The fragmentation of discs necessary for repeated
transplantation means that they are in a state of continuous regeneration.
If they are returned to a late larva then the implants do differentiate
as the host undergoes metamorphosis. In most cases the structures formed
are appropriate to the original character of the disc, but sometimes
they are appropriate to a different disc. For example, tissue from a
leg disc may produce wing structures (Hadorn 1968). Recent work has
shown that ectopic expression of the transcription factor gene vestigial
can drive transdetermination of other discs to wing phenotype (Maves
and Schubiger 2003). In all these examples the transformation occurs in the course of regeneration, which somehow destabilises the differentiated or committed cell types. In most cases the original and final tissue type are developmentally related in some way, for example being formed as adjacent regions from one sheet of cells in the embryo. It is now known that the specification of tissue types during embryonic development depends on the action of inducing factors secreted from signalling centres. These form gradients of concentration across the target tissue and different concentrations can activate the expression of different transcription factors (Figure 1). The developmental specification of each region depends on the combination of transcription factors that are expressed. The nature of the signalling process means that, in general this combination will be more similar for regions that are adjacent or close together. Transdifferentiation events will occur if the combination of transcription factors is altered in the course of regeneration. This may happen because of somatic mutation, or the operation of inducing factors in the immediate environment, combined with destabilisation of the states of gene activity due to chromatin decondensation. It need only occur in one or a few cells, and if the new tissue type has a growth advantage over the old, it will expand to become a macroscopic focus.
Fig.1 Model for metaplasia. Reproduced from Slack and Tosh (2001) Transdifferentiation events resemble homeotic mutations, which are well known particularly in Drosophila, and have long been known to bring about changes in the combinations of transcription factors that specify particular body parts such as imaginal disc identity. The principle of homeotic mutation is explained in Figure 2. In this organism the head and three body segments arose in the embryo from a common cell sheet through the operation of a morphogen which has high concentration at the future posterior end and low concentration at the future anterior end. Three regulatory genes were activated at different concentration thresholds such that the embryo consists of four territories of which the gene activity codes, from anterior to posterior, are 000, 001, 011, 111, where 0 indicates a gene is off and 1 indicates it is on. If the second gene is mutated to inactivity then the codes will become 000, 001, 001, 101, which will lead to a homeotic switch of the third segment to the character of the second. Conversely if the second gene is caused to be expressed all over the embryo, then the codes will be 010, 011, 011, 111, and this will lead the second segment to develop with the character of the normal third segment.
Fig.2 How homeotic mutations can alter the character of body parts (Reproduced from Slack 2005) Transdifferentiation of transplanted stem cellsIn the years 1998-2002 there were many reports of differentiated body tissues being populated by cells from bone marrow grafts (reviewed Tosh and Slack 2002; Raff 2003; Wagers 2004). These experiments were mostly conducted in mice. In a standard protocol the host mouse receives a dose of radiation which destroys its own bone marrow. This will inevitably lead to death through inability to replace the cells of the blood and immune system. If bone marrow from a healthy donor is injected, it will colonise the marrow of the host and repopulate the blood and immune system with healthy cells. The availability of donor mice carrying visible genetic markers, such as beta-galactosidase or green fluorescent protein, led to the discovery that a small minority of cells in tissues throughout the body of the hosts were of donor origin. The tissues concerned included the brain, skeletal and cardiac muscle, and epithelial tissues such as liver and kidney. Bone marrow contains many cell types, but the cell that is ultimately responsible for long term repopulation of the blood and immune system is the haematopoietic stem cell (HSC). Similar results were obtained when purified preparations of HSCs were used for the graft suggesting that these cells were capable of transdifferentiating into a wide variety of other tissue types. This led to suggestions that it might be possible to cure degenerative diseases of the heart or brain simply by grafting of stem cells. It was also suggested that there is a continuous turnover of cells in the entire body, fed ultimately by the HSCs of the bone marrow. However further investigation did not bear out these speculations. In some cases it was shown that the donor cells had fused with host cells and so the genetic marker of the donor was still displayed in the fused cells. In some cases the donor cells were simply lodged in the host tissues but the cells had not actually transdifferentiated. Different investigators also obtained very different results with similar protocols. One difficulty is the radiation normally given to the hosts. This causes a lot of tissue damage throughout the body, creating a lot of local cell death, wound healing and regeneration processes. This is fertile ground for sequestration of the donor cells or for cell fusion. It is possible that in rare cases the marker gene alone might be taken up by host cells. Given all these problems the consensus now is that transdifferention of grafted HSCs does not occur, or if it does occur it is on a very small scale. Use of transdifferentiation for therapy.Although transdifferentiation is a very rare event in nature, understanding
how it happens enables us to design potential methods for causing it
to happen by activating or inhibiting the expression of certain key
transcription factors in the target tissue. Although this is still in
the range of basic research we can imagine:
|
