The basics about research in the Oshimori Lab

Tissue-resident (adult) stem cells

A single fertilized egg gives rise to all the cell types in the body. When individual tissues form, they set aside reservoirs of stem cells that become more restricted in their lineage options. These fascinating cells live in the tissue for the entire life. Tissue-resident stem cells make and replenish the differentiated cells within the tissue that are continually lost during tissue turnover or following injury.

The stem cell niche

Stem cells have the remarkable capacity to divide and maintain the undifferentiated state, a process called self-renewal. Stem cells are often used sparingly to give rise to rapidly proliferating but transient cells that perform tissue regeneration. Stem cells that spend much of their time in a quiescent state are protected from unnecessary cell divisions that could lead to cancer.

The skin epidermis: an excellent model for stem cell research

We are interested in how stem cells maintain normal tissues (homeostasis) and repair wounds after injury, and how the dysregulation of stem cells cause tumor initiation and progression. We use skin epidermis as a model system to understand the mechanisms that regulate stem cells in mammalian tissues.

The skin protects our body from physical, chemical, and biological insults. The skin also retains body fluids, regulates body temperature, and senses the outside world. For these functions, mammalian skin develops the epidermis and its appendages, including hair follicles and sweat glands. The epidermis is a multilayered epithelial tissue mainly composed of keratinocytes. The basal layer, the epidermis’ innermost layer, harbors stem cells, which continually self-renew and produce differentiating progeny cells, replenishing the epidermis by pushing older cells up toward the surface of the skin.

Hair follicles are tubular invaginations of the epidermis into the dermis and produce hair fibers in a cyclic manner. Hair follicles harbor dedicated stem cells in the deepest portion of the permanent follicle called the bulge. Bulge stem cells represent the most quiescent and long-lived epithelial cells in the skin.

Secreted factors emanating from the bulge niche regulate the quiescence and activated states of stem cells. For example, bone morphogenetic proteins (BMPs) induce the quiescent state. In contrast, fibroblast growth factors (FGFs), Wnt, and transforming growth factor b (TGF-b) function as activation cues that overcome the inhibitory signals and launch a new hair cycle.

TGF-b: a critical mediator of stem cell biology

For regenerating tissues, stem cells receive external cues from their microenvironment and integrate signals with their intrinsic abilities to self-renew and differentiate. Therefore, elucidating signaling molecules that help stem cells detect tissue contexts is essential for understanding stem cell biology.

Transforming growth factor β (TGF-β), an evolutionally conserved cytokine, constitutes an integral component of the cellular crosstalk between stem cells and the niches. All TGF-βs (TGF-β1, TGF-β2 and TGF-β3) are translated as larger polypeptides whose amino-terminal prodomains are cleaved but required their dimerization. Moreover, the noncovalent association between TGF-β and the prodomain persists, and thus TGF-β is secreted as a latent (i.e., biologically inactive) complex and sequestered in the extracellular matrix. The release of active TGF-β is regulated by diverse mechanisms, including extracellular protease-mediated cleavage, low pH, reactive oxygen species, and specific integrins that can bind to the Arg-Gly-Asp (RGD) sequence within the prodomain.

TGF-β ligands bind to the type I and type II TGF-β receptors (TβRI and TβRII) and bring the two receptor serine-threonine kinases together. Upon TGF-β–mediated activation, TβRII phosphorylates TβRI, which then propagates the signal by phosphorylating intracellular effector substrates. The canonical downstream effectors are SMAD2 and SMAD3. Phosphorylated SMAD2/3 can translocate to the nucleus and bind to the specific DNA sequence called SMAD-binding elements (SBEs) with SMAD4. They can either activate or repress transcription through their interaction with co-activators such as p300/CBP histone acetyl-transferases (HATs) or co-repressors such as histone deacetylases (HDACs) or the SWI/SNF chromatin-remodeling complex. Additionally, TGF-β activates SMAD-independent pathways such as PI3K/Akt, MAPK (ERK, JNK, and p38) and NF-κB pathways.

At high levels, TGF-β induces a reversible proliferation arrest in epithelial cells. TGF-β’s cytostatic function is crucial for stem cells because many adult stem cells must survive for months in a quiescent state. In contrast to terminal differentiation and senescence, quiescent stem cells can re-enter the cell cycle in response to specific environmental cues. TGF-β in balancing quiescent and proliferative states may be critical not only to preserve tissue-resident stem cells but also to respond quickly to tissue demand.

Stem cells and cancer

Cancers develop from normal cells that eventually gain the ability to proliferate aberrantly and become malignant. Tissue-resident stem cells may be preferential targets for oncogenic transformation because they are long-lived and have time to accumulate the myriad of mutations that will lead to cancer. Using mouse models of solid tumors, researchers have shown that oncogenic mutations in stem cells effectively induce tumor initiation compared with non-stem cells, including actively-dividing, transit-amplifying cell populations.

Squamous cell carcinoma (SCC): a deadly skin cancer arising from keratinocytes

In the course of tumor development, CSCs and their niches are thought to evolve through reciprocal interactions. However, the mechanisms by which the CSC–niche relationship emerges during malignant transformation is poorly understood. An excellent model for investigating cellular crosstalk and molecular changes during early tumor development is squamous cell carcinoma (SCC) of the skin, the second most prevalent cancer worldwide with an increasing incidence rate (>250% increase between 1980 to 2010) and a significant risk of metastasis. Although skin SCC is often curable by surgical excision, ~8% of patients show relapse and 5-20% present metastasis within five years, and of these cases, the 5-year survival rate is only 25-35%. Therefore, an improved mechanistic understanding of early-stage SCC is urgently needed to identify targets for prognostic assessment of at-risk populations and early therapeutic intervention.

SCCs exhibit high rates of tumor recurrence following chemotherapy and radiotherapy. SCCs have stem cell populations, which represent ~1%-5% of the tumor cells and reside at the tumor–stroma interface. They are typified by elevated expression of specific integrins and other stem cell markers, including CD34, CD44, and SOX2. They also express vascular endothelial growth factors (VEGFs), suggestive of enrichment at the perivascular areas. Interestingly, heterogeneity in proliferation rates exists within these CSCs, i.e., faster-cycling and slow-cycling populations.

TGF-b in cancer

The role of TGF-β during tumorigenesis is complex. TGF-β controls tissue homeostasis and suppresses tumor initiation by regulating cell proliferation, differentiation, and apoptosis. Consistent with this, the loss of TGF-β signaling by conditional ablation of TβRII predisposes epithelial tissues to cancer. However, while elevated TGF-β signaling in the skin decreases the formation of benign tumors, TGF-β increases the rate of malignant conversion from premalignant tumors to malignant SCC and promotes metastatic progression. Therefore, TGF-β play a dual role in tumor development: TGF-β functions as a tumor suppressor in early stage tumors while paradoxically acting as a tumor promoter in later stages. Because TGF-β exerts a tumor-supportive role by regulating non-tumor cells in the stroma, such as the activation of cancer-associated fibroblasts and immunosuppression, the precise role of TGF-β signaling in malignant progression of tumor cells remain largely unknown. A major technical obstacle exists in the conventional gene knockout approach (i.e., the lack of TGF-β signaling throughout tumorigenesis), which does not allow for examining the tumor cell-intrinsic effects of TGF-β in tumor progression. To address the role of TGF-β signaling in tumor cells, we need a system to detect and modify TGF-β signaling after the tumor has already formed. Recently, we generated an innovative mouse model of SCC that allows us to study a significant cancer biological problem in a physiological system with the highest level of tissue and cell-type specificity.

Ultrasound-guided in utero microinjection of lentiviral vectors

Both tumor-initiating stem cells and cancer stem cells are embedded in a specific tissue architecture, which is continually evolving during tumor initiation, promotion, and malignant progression. To understand cellular crosstalk and molecular pathways that regulate tumor-initiating cells during early tumor development, we need a system to induce tumorigenesis from a native stem cell population at a clonal level, detect activation of signaling pathways in situ, and analyze their functional properties in developing tumors.

We generated a mouse model of de novo SCC by applying an epidermis-specific lentiviral gene delivery system. We designed lentiviral vectors encoding a tetracycline-inducible transcriptional activator (rtTA), a Cre recombinase, and a TGF-β signaling fluorescent reporter driven by an SBE-regulated promoter. By injecting the lentiviral particles into the amniotic fluid surrounding mouse embryos in utero, epidermal progenitor cells of TetO-Hras, Rosa26-LSL-EYFP transgenic mouse are transduced sparsely with the lentiviral vectors, therefore enabling to assess clonal behaviors of tumor-initiating cells.

Postnatally, transduced epidermal stem cells and their progeny are visualized through yellow fluorescent protein (YFP) expression induced by lentiviral Cre. Then, to initiate tumor formation, we activate oncogenic HRAS expression in YFP⁺ cells by lentiviral rtTA in a doxycycline-dependent manner. Transformed YFP⁺ cells initially form SCC in situ (defined as a tumor that is limited to the epidermis and has not invaded into the dermis) and often progress to well-differentiated SCC (a tumor that closely resembles normal epidermal differentiation patterns), and eventually to poorly-differentiated, invasive SCC (a biologically aggressive tumor that is prone to recur and metastasize).

An in vivo reporter to detect TGF-b–responding tumor cells

The fluorescent reporter in the lentiviral vector enables us to detect the activity of TGF-β–SMAD2/3 signaling pathway in vivo. We often observed histopathologically distinct tumor areas manifesting well-differentiated and invasive SCC in the same tumor, and lentiviral fluorescent reporter illuminated TGF-β–responding tumor cells at the tumor–stroma interface of invasive SCC. Notably, the frequency of TGF-β–responding tumor cells was spatially associated with localized TGF-β ligand expression in the adjacent stroma. Consistent with cytostatic effects of TGF-β, TGF-β reporter⁺ cells were a slow-cycling tumor cell population.

Taking advantage of the fluorescent reporter signal, we fractionated tumor basal cells into TGF-β–responding and the non-responding populations by fluorescently-activated cell sorting (FACS). The TGF-β–responding population frequently expressed bulge stem cell marker, CD34. Moreover, these cells exhibited higher colony-forming efficiency in vitro and higher tumor-propagating capacity in transplantation assays in vivo. Thus, tumor-initiating stem cells are enriched in the TGF-β–responding subset of SCC basal cells.

A lineage tracing system to study the fate of TGF-b–responding tumor cells

Lineage tracing is a powerful approach for delineating the fate of stem cells and their progeny with minimal disturbance of their physiological function in vivo. To study the role of TGF-β signaling in SCC development, we generated a next-generation reporter that induces a tamoxifen-inducible Cre (CreER) explicitly in TGF-β–responding tumor cells. Upon tamoxifen administration, Rosa-YFP marks TGF-β–responding tumor cells as well as their progeny. This system allowed us to examine how TGF-β–responding cells behave during tumor development and drug treatment.

We showed that TGF-β–responding tumor cells give rise to invasive areas of SCC. The progenies of TGF-β–responding tumor cells lose the normal differentiation marker expression. Moreover, the progenies tended to be more scattered at the tumor–stroma interface proximity to tumor vasculatures and exhibited phenotypes resembling epithelial–mesenchymal transition (EMT). Our data suggest that TGF-β is involved in a non-genetic mechanism of tumor heterogeneity that underlies the malignant progression of SCC.

It has long been suggested that slower-cycling, tumor-initiating stem cells might be refractory to conventional chemotherapy. One of the most widely used anti-cancer drugs, cisplatin [cis-diamminedichloroplatinum (II)], is the standard chemotherapy for head & neck SCC and advanced skin SCC. However, tumor recurrence remains a major clinical problem. Surprisingly, TGF-β–responding cells escaped cisplatin-induced apoptosis in shrinking tumors. Moreover, lineage tracing studies demonstrated that recurrent tumors were disproportionally occupied by the progeny of TGF-β–responding tumor cells, suggesting that TGF-β signaling is a critical determinant for the emergence of drug-resistant cancer stem cells (CSCs). Therefore, understanding the mechanisms by which TGF-β endows tumor cells with malignant properties could potentially offer a novel strategy of CSC-targeted therapies.

Transcriptome analysis of TGF-b–responding tumor cells

The ability of TGF-β signaling to confer enhanced survival to tumor-initiating stem cells of chemotherapy-treated tumors was consistent with the cancer stem cell hypothesis for tumor recurrence. To gain insight into drug resistance, we purified TGF-β–responding tumor cells by FACS and analyzed their transcriptional profile by RNA sequencing (RNA-seq). As expected, they upregulated known CSC signature genes, such as Sox2 (a transcription factor essential for maintaining self-renewal of stem cells) and Vegfa (the vascular endothelial growth factor that promotes tumor angiogenesis).

Antioxidant responses, glutathione metabolism, and the transcription factor NRF2

Gene ontology (GO) and molecular pathway analyses provided insights into biological processes activated in TGF-β–responding tumor cells. Redox—reduction and oxidation—genes and molecular pathways involved glutathione metabolism, and antioxidant responses were surfaced from these analyses. Glutathione (GSH) is the most abundant intracellular antioxidant in animal cells. It involves two critical metabolic processes, 1) reduction reaction, which prevents damage from reactive oxygen species (ROS) by exhausting ROS through the conversion of reduced GSH to its oxidized state (glutathione disulfide, GSSG), and 2) conjugation reaction regulated by glutathione S-transferases (GSTs), which is known to metabolize chemotherapeutic drugs including cisplatin.

The myriad of glutathione metabolism genes upregulated in TGF-β–responding tumor cells are transcriptional targets of NRF2, the master transcription factor for antioxidant responses. The activity of NRF2 is tightly controlled at a protein level by the KEAP1–mediated ubiquitin–proteasomal degradation pathway. We found that TGF-β–responding tumor cells exhibited stabilized NRF2 in the nucleus. Indeed, shRNA-mediated NRF2 gene knockdown sensitized TGF-β–responding tumor cells to chemotherapy-induced apoptosis.