Cell Reprogramming for Regenerative Medicine (2)
Modeling of diseases in vitro
(Continuation, the beginning of the article is here.)
Researchers have already begun work on creating experimental models by reprogramming cells of patients with a wide range of diseases, including cystic fibrosis, Huntington's disease, Parkinson's disease, sickle cell anemia, congenital dyskeratosis, familial amyotrophic lateral sclerosis. This list is growing rapidly (Grskovic et al., 2011; Park et al., 2008b; Dimos et al., 2008; Mali et al., 2008; Somers et al., 2010; Ghodsizadeh et al., 2010; Agarwal et al., 2010). Working with such models will allow us to study not only the mechanisms of disease development, but also the relationship between genotype and phenotype.
For example, the study of hereditary long QT syndrome in vitro was hampered by the inability to obtain human cardiomyocytes carrying a causal mutation (Behr et al., 2008). As part of a recent work, scientists obtained iPSCs of patients with type I monogenic long QT syndrome, differentiated them into cardiomyocytes and analyzed their characteristic electrophysiological parameters (Moretti et al., 2010). Compared with the cells of healthy people, the cells of patients demonstrated a longer action potential and abnormal localization of proteins. Based on these data, the authors determined the mechanism of the disease development as dominant-negative. In an article published a little later, the researchers describe the results of studying another variant of the hereditary long QT syndrome, the cause of which is a mutation in another gene (Itzhaki et al., 2011). The authors identified additional electrophysiological features of patients' cells associated with the disease. In addition, the authors conducted a small screening, the purpose of which was to assess the ability of chemical compounds to facilitate the manifestations of the disease. The existence of two such models makes it possible to directly compare cellular phenotypes caused by different genotypes in the same disease, and also, possibly, will increase the effectiveness of its treatment (Figure 2).
Figure 2. Opportunities provided by disease models created by reprogramming cells.
The results of dozens of similar studies have already been published and summarized in recent review articles (Grskovic et al., 2011; Tiscornia et al., 2011; Unternaehrer and Daley, 2011). However, do not forget that the methods of using iPSCs for modeling diseases are still in their infancy and researchers working in this field face serious difficulties. The following are the issues that should be solved by specialists working on models of any diseases.
Protocols of direct differentiation and cell cultureWork on methods for converting pluripotent cells into cells of a certain type has been going on for many years.
Some fully functional highly specialized cells, including motor neurons and cardiomyocytes, have already been obtained using protocols that repeat the mechanisms involved in the development of the embryo. However, for other "long-awaited" cells, such as pancreatic beta cells and hematopoietic cells, only similarities were obtained at best, the phenotype of which differs significantly from the original (Kroon et al., 2008; Zhang et al., 2009). The main reason for this is the lack of information about the conditions guiding the differentiation of these cells during embryo development.
Determining the authenticity of "target cells"The tissue belonging of cells is determined by their phenotype, consisting of many components.
To work with cells of one type or another obtained from iPSCs, specialists must first make sure that the product grown in vitro is identical to its prototype in vivo. This identity assessment should be carried out at different levels, it should include an analysis of gene expression, chromatin state and functionality. Transcription activity and methylation level are evaluated equally for all cell types, whereas specific markers of type affiliation, as well as methods for analyzing tissue-specific functions should be selected individually for each type of target cells.
Identification of adequate disease-specific cellular phenotypesIn some cases, cells obtained from iPSCs of patients with various diseases demonstrate clear predictable phenotypes, such as the electrophysiological abnormalities described above in cardiomyocytes.
In other diseases, the result of the analysis for the phenotype of the disease may be ambiguous or may indicate the absence of a statistically significant difference between the lines of disease-specific and normal cells. For example, a recently published article describes a comparison of iPSC lines isolated from patients with idiopathic Parkinson's disease and healthy people in the control group (Soldner et al., 2009; Hargus et al., 2010). The authors planned to compare the ability of iPSCs of healthy people and patients to form dopaminergic neurons when using the direct differentiation protocol, as well as the functioning of such neurons during their transplantation into various animal models of Parkinson's disease. In practice, the difference between disease-specific and normal neurons was revealed only in a mouse model in one of three behavioral tests.
Such cases should remind researchers of their responsibility for confirming that the "disease-specific phenotypes" they have created correspond to the biological features of the disease in question.
Identification of diseases for which working with iPSC-based models is suitableThe example of idiopathic (caused by an unknown cause) Parkinson's disease is a warning that modeling diseases using iPSC is more informative in some cases than in others.
Perhaps reprogramming is still too little–tested a tool in order to predict which diseases modeling is effective and which are not. However, this issue should be resolved in the future.
Identification of diseases most suitable for working with models created from embryonic stem cellsDisease-specific embryonic stem cell lines obtained during preimplantation genetic diagnostics are better suited for studying certain conditions than iPSC lines.
One example is the syndrome of the brittle X chromosome, which is characterized by abnormal suppression of the activity of the FMR1 gene during development. Due to the inability to reactivate the mutant locus during reprogramming (Urbach et al., 2010), disease-specific iPSCs do not express the FMR1 gene. These cells can differentiate into neurons with an inactivated FMR1 gene (Sheridan et al., 2011), but they do not allow us to study the mechanisms that provide pathological suppression of gene activity.
Reprogramming does not perfectly "reset" the epigenome of cellsSeveral studies have been devoted to comparing the epigenetic status of iPSCs and embryonic stem cells.
According to the global methylation profile, iPSCs are much closer to embryonic cells than to tissue cells of their origin (Doi et al., 2009; Lister et al., 2011), however, the epigenomes of iPSCs and ESCs are not completely identical.
It has been found that certain classes of epigenetic markers avoid the "data zeroing" that occurs during reprogramming. When trying to differentiate iPSCs isolated from different tissues into cells of different types, it was noticed that reprogrammed cells differentiate more easily into cells of tissues belonging to the same type to which the original cells belonged. In some cases, iPSCs even retain residual methylation, reflecting their original tissue affiliation. These data indicate the presence of epigenetic memory in iPSCs, the essence of which is that a small number of epigenetic labels are not eliminated during reprogramming, which apparently happens randomly (Kim et al., 2010, 2011). However, the researchers also noticed that with prolonged cultivation, the methylation profile of iPSCs gradually approaches the methylation profile of ESCs (Nishino et al., 2011). Other scientists have shown that certain regions of chromosomes located in the immediate vicinity of telomeres and centromeres may be particularly resistant to the removal of epigenetic labels during reprogramming and that the DNA methylation features of these regions may persist even during differentiation (Lister et al., 2011).
Causes of epigenetic variability in vivo: environment and stochasticityExperts have long known that even transient effects of environmental factors often lead to changes in the epigenome that have a lasting effect on cell behavior due to changes in transcriptional activity (Bell and Spector, 2011; Waterland and Jirtle, 2004; McGowan et al., 2009; Anway et al., 2006).
Stochastic processes also contribute to the variability of the organization of the epigenome. The estimated reproduction accuracy of inherited specific methylation in humans is 90-98% (Ushijima et al., 2003; Genereux et al., 2005). However, it is obvious that even the existing small error with each cell division makes changes in the cell's methylome at a much higher rate than the rate at which DNA sequence mutations can accumulate in principle. Unlike predicted epigenetic modifications under the influence of environmental factors, this type of changes in DNA methylation occurs randomly.
Simultaneously with the appearance of environmental factors and stochastic changes in the human epigenome, variations slowly accumulate, reflecting the conditions of his life (Wong et al., 2010). Some people may also be genetically predisposed to more or less pronounced periodic changes in the epigenome (Bjornsson et al., 2008; Feinberg and Irizarry, 2010). Taking into account the fact that a certain part of these variations affects the expression of genes, an individual epigenome determines the reactions of cells to certain influences and, accordingly, can cause a predisposition to diseases or good health.
End: Genomics, Epigenomics and external influencesPortal "Eternal youth" http://vechnayamolodost.ru