PI – Dmitri Papatsenko
Researchers – Ivan Kulakovskiy, Artem Kasianov, Dmitry Shcherbinin
PhD students – Anna Shmelkova, Alexey Samosiuk
Dmitriy Papatsenko laboratory explores the following main research directions:
1. Variation of gene expression in adult and embryonic stem cells and stem cell subpopulations.
Medical use of stem cells is limited due to their rather unpredictable behavior – while the majority of embryonic stem cells undergoes self-renewal, some cells may spontaneously commit differentiation or even become cancer cells. Explanation of such behavior varies from stochastic variation of gene expression to existence of alternative deterministic states in the stem cell populations, which prime the cells to follow certain differentiation or transformation paths.
Using analysis of variation of gene expression in single embryonic stem cells (ESC) we have previously shown existence of at least two subpopulations in ESC culture grown on serum supplemented with LIF (leukemia inhibitory factor). The identified ESC subpopulations were characterized by specific gene expression patterns. According to the proposed model (see Figure 1), the two subpopulations exchange with each other, so there is an equilibrium, supporting the composition of ESC populations grown on serum + LIF media.
Figure 1. Stalled Differentiation and State Exchange Model for Self-Renewal
Relative positions of the detected ICM and transient pluripotent states are shown in the context of mouse embryo development, the color bars below display the relative expression levels of uniformly (OCT4 and SOX2) and differentially (TBX3 and NANOG) expressed genes between the two pluripotent states. The color bar on the right demonstrates higher absolute expression levels of core pluripotency factors in the 2i state. Self-compensatory network motifs, such as the one at the bottom left (iFFL-FB), may be responsible for the stabilization of ICM or other pluripotent states. The transient state is more mature than the ICM state, and it is more similar to epiblasts. ESC self-renewal may correspond to a dedifferentiation (blue arrow), which occurs when differentiation is stalled and the cells begin to roll back from the transient state to the ICM state. Such a dynamic exchange between the two states may ensure maintenance of pluripotent mESCs.
(from: Papatsenko D., et al., Stem Cell Reports. 2015 Aug 11;5(2):207-20).
Currently, we investigate variation of gene expression using integrated analysis of single cell data, available for embryonic and adult stem cells with the aim to identify genes preferentially expressed in the subpopulations, to establish the potential mechanisms of transitions between the subpopulations and to identify gene networks, which stabilize pluripotency states, responsible for the formation of the subpopulations.
Investigation of stable pluripotency states and mechanisms of self-renewal in ESC or induced pluripotent stem cells (iPSC) requires deep understanding of underlying regulatory mechanisms. In the case of ESC and iPSC regulation is achieved by tightly coordinated action of several network layers, including signal transduction, regulation of transcription and regulation at the level of epigenetics. Transcriptional layer, represented, in particular by the core pluripotency gene network (transcription factors Oct4, Sox2, Nanog) is especially important as it is believed that it may serve as the central processing unit (see Figure 2).
Figure 2. Hierarchical organization of pluripotency gene regulatory network (PGRN)
Signaling pathways represent the first hierarchical level (in green). Signaling pathway mediators (in yellow), including transcription factors, which enter nucleus upon phosphorylation represent the second hierarchical level. Components of the transcriptional pluripotency network (shades of red) represent the third level. Note that within this level some transcription factors such as Oct4, Sox2, and c-Myc occupy higher position, while factors that are less critical for pluripotency occupy lower positions. Lowest level of the pluripotency network (in violet) include components of epigenetic complexes involved in chromatin modification, DNA methylation and regulation at posttranscriptional levels, as indicated. Red and blue arrows show positive and negative interactions correspondingly; blue T-arrows show targets of inhibitors for “2i” conditions. Feedback control loops are not shown in this figure (manuscript in submission).
We have previously reconstructed general hierarchy of pluripotency gene networks by browsing data available in the literature, and our next goal is to identify exact order and direction of regulatory connections between the most essential regulatory factors, highly expressed in ESCs or iPSCs or at the early stages of differentiation of these cells.
Our major goal is to integrate various kinds of data, including in vivo binding data (ChIP) and expression time series (and other) data available for ESCs and iPSCs from human and mouse. Comparative analysis of binding sites for transcription factors and the results of knock-downs (or overexpression) of the same factors should allow reconstruction of high quality gene regulatory networks and provide grounds for exploration of dynamic behavior of these networks under changing conditions.
Integrated analysis of single cell data (project #1) as well as ChIP and bulk gene expression data (project #2) should result in reconstruction of regulatory gene networks acting in ESCs and iPSCs (data from early mouse embryos will be taken into consideration as well). We are interested to understand how these pluripotency gene networks carry out their functions in stem cells in vivo and in vitro, how the networks would respond to small changes of environment (addition of growth factors etc to media – extrinsic factors) or deregulation of certain network components (knockdowns or overexpression- intrinsic factors).
In order to understand the principles of functioning of the reconstructed regulatory networks, we are going to built quantitative models, describing dynamics of gene regulation in ESCs and iPSCs. The quantitative models should predict the network responses to changing inputs (extrinsic and intrinsic factors) and sufficiently describe the majority of available experimental data related to loss of function or gain of function data for major pluripotency genes.
Predictive quantitative models describing cell fates and their transitions should help to identify very effective molecular switches, which allow to control the stem cell behavior in a desired way with minimal intervention. The identified molecular switches (small sets of regulatory factors) may be utilized in many practical applications, including cellular reprogramming using predicted sets of factors, efficient directed differentiation, production of high-quality human ESCs and iPSCs, suitable for regenerative medicine (currently medical use of these cells is prohibited).
4. Pluripotency maintenance and homeostasis in mammalian germ line stem cells.
5. Transcriptional regulation of pluripotency genes in embryonic stem cells.
6. Effect of polycomb repression complexes on self-renewal of epidermal stem cells.
7. Oxidative stress in the regulation of normal and neoplastic hematopoiesis.