 
|
 |
Tumor cells that exit the primary tumor and are able to enter the arterial or lymphatic system can reach distant organs physically
disconnected from the primary tumor mass. Once in the vascular bed of these targets organs, cells can remain within the
microvasculature or extravasate (exit the blood vessels) and lodge in the target organ parenchyma. This process is initiated once
cells that have acquired malignant proterties and it is known as the metastatic cascade (Fig1). While in many
cases cells can resume growth immediately, in some situations cells may enter a state of dormancy. It appears that this stage is
characterized by a G0/G1 arrest. What mechanisms determine that solitary cells (or small groups of cells) that were competent for
survival and proliferation in the primary site, suddenly become growth arrested and dormant in target organs?(Fig1).
| Click image for larger size |
|
|
Fig1. Tumor formation and progression to metastasis. (A), Carcinoma forms in the epithelial compartment, which has a distinct polarized
organization maintained through cadherin based cell-cell junctions and integrin-based cell-matrix attachments. The cells rest on a
basement membrane. Nutrients are provided by blood vessels in the stromal compartment (Normal tissue). (B), Mutations and accumulating
epigenetic changes alter cell morphology and cause loss of polarized architecture. The proliferative compartment increases. Proteases
may, or may not be expressed, but the basement membrane remains intact (In situ carcinoma). (C), Further genetic or epigenetic events
lead to the acquisition of motile and invasive properties (strong, self-generated mitogenic signaling, active integrins and
proteases), the cells degrade the basement membrane, invade the underlying stroma. The heterogeneity within the tumor increases.
Invading tumor cells interact with fibroblasts (brown) or immune cells (blue and pink) and matrix component different from
those in basement membrane. Angiogenesis may occur (not shown) and tumor cells (in cooperation with stromal cells) degrade
the matrix and the vascular walls (yellow) and disseminate through arterial or lymphatic routes (Invasive carcinoma), (D).
The cells arrest in lymph nodes, bone marrow, or in blood vessels of target organs, where they extravasate into the organ
parenchyma by interacting with the endothelium and degrading the vessel wal (E). In the new environment tumor cells can
rapidly form growing, detectable metastases or they can remain dormant for variable time periods (months to decades)
until new signals are generated that activate the cells to resume their growth (E).
|
Our long-term goal is to identify the mechanisms that regulate metastatic growth. In particular we are interested in the mechanisms that favor dormancy of in situ or
disseminated tumor cells. More than half of cancer patients will die from metastatic disease that develops months, years or even decades after primary tumor removal.
It appears that during these periods disseminated cells are in a dormant, non-proliferative state. Clearly, forcing proliferating cells into dormancy or maintaining cells
dormant would be highly beneficial for patients. However, the lack of models for identification and isolation of dormant cells and the scarcity of these cells has hampered
studies on the mechanisms governing dormancy.
We employ a model of cancer dormancy in vivo using a human squamous carcinoma (HEp3) and nude mice or chick embryo animal xenograft models (Ossowski et al., 1980, 1987, 1991).
In HEp3 cells, urokinase receptor (uPAR), which is over-expressed in most cancers, interacts with and activates a5b1-integrin, recruiting and activating the epidermal growth factor receptor
(EGFR) (Fig2). Activation of EGFR results in strong activation of the extracellular regulated kinase (ERK) and cancer cell proliferation in vivo. Concomitantly, the uPAR-α5β1-EGFR complex
inhibits the p38 growth-suppressive (stress) pathway; such inhibition results in increased mitogenesis. Disruption of the uPAR-complex results in ERK inhibition and activation of p38; under
such conditions G0/G1 arrest is induced and dormancy is maintained (Kook et al., 1994, Yu et al., 1997, Aguirre-Ghiso et al., 1999, 2001, 2002, 2003, 2004, Liu et al 2002). Further, mechanistic
analyses revealed that p38 activation establishes a negative feedback loop on ERK activation and either genetic or pharmacological inhibition of p38 was sufficient to restore ERK activation
and interrupt tumor dormancy (2, 3). Together these results pointed to p38 stress-activated signaling as an important pathway capable of reversing malignancy through the induction of dormancy.
Interestingly, p38 signaling induced growth arrest without increasing apoptosis (2).
|
|
Fig2. uPAR, the ERK/p38 ratio and tumor dormancy. (Left) In uPAR-rich cells uPA-bound uPAR frequently interacts with α5β1 integrin, causing its activation.
This leads to integrin-dependent recruitment of FAK/EGFR complex that results in activation of Ras-ERK signaling. Concomitantly this complex, maintains Cdc42
inactive and prevents p38 activation. This results in a mitogenic high ERK/p38 signaling ratio. (Right) Downregulation of uPAR or blocking of α5β1 function
results in integrin inactivation, disassembly of the complex, inactivation of its intracellular signaling components and reduced ERK activation. In contrast,
Cdc42 becomes activated and induces p38 activation. This results in a growth arrest-inducing low ERK/p38 ratio, which forces tumorigenic cells into dormancy.
|
What is cancer dormancy and why is it important to understand the biology behind it?
Dormancy is defined as the ability of cells or organisms to be in a condition of biological rest or suspended animation during which they are not active but capable of becoming active.
For example, C. elegans can enter a larval stage named dauer, which is triggered by stress signals such as nutritional deprivation and these organisms undergo a protracted developmental
arrest until conditions become propitious for growth. Another good example where environmental cues can trigger a stress resistant behavior that postpones development until the conditions
are appropriate is the dormancy of plant seeds. Only when environmental conditions are appropriate seeds will emerge from dormancy and germinate. An analogous behavior is observed in
dormancy of recurrent in situ or disseminated tumors.
Clinical evidence indicates that patients with minimal residual disease harbor dormant cells that are not dividing but still are able to do so after periods that could last for years or even decades.
These cells in addition are found to be refractory to chemotherapy regimens. This could be due to slow or absent proliferation and/or to mechanisms that provide a high tolerance to stress.
Since, understanding the biology of dormancy and the chemo-resistance of the residual metastatic disease remains a formidable challenge, gaining insight into these mechanisms represents
a most urgent aspect of study. New therapies to stop metastatic growth will be possible only with a better understanding of the biology behind the induction of dormancy and the drug-resistance
of dormant tumor cells. The significance of our studies rests on the fact that understanding the mechanisms regulating induction of dormancy and survival of dormant cells is essential to identify
new treatments tailored to (i) stop metastatic growth by inducing dormancy and subsequently fully eradicate in situ or disseminated dormant cancer cells by rendering them susceptible to chemotherapy.
Converting progressive cancer into an asymptomatic-chronic condition and subsequently eradicating minimal residual disease would most certainly lead to life saving and/or life prolonging therapies for patients.
To better understand the gene programs that metastatic cells regulate to grow or enter a protracted state of dormancy new tools were required that allowed us to follow
these programs in vivo. For this purpose we have developed a system that allows us to monitor neoplastic cell signaling signatures in vivo: We have found that
measurement of ERK or p38 signaling serves to indicate the behavior of different human cancers in vivo (Aguirre-Ghiso et al., 2001, 2003, 2004). On the strength
of these findings, we hypothesized that a reporter system in which GFP expression could be controlled by ERK or p38 signaling would allow monitoring of these
pathways in tumor cells in vivo. We have developed such a system in which ERK or p38 activation regulates GFP expression through the activation of a GAL4-Elk
or GAL4-CHOP trans-activator, respectively (Aguirre-Ghiso et al., 2004). This system has aided in the spatial-temporal monitoring of these pathways in primary
and metastatic sites (for example see Fig3). Using this novel reporter system we were able to examine the fate of these cells in culture, in primary tumors and
in spontaneous metastasis in chick embryos and nude mice. In culture GFP-level was directly proportional to the previously established levels of ERK or p38
activation. In contrast, during the first 24hrs of in vivo inoculation, both the tumorigenic and the dormant cells strongly activated the p38 pathway. However, in
the tumorigenic cells p38 activity was rapidly silenced, correcting the ERK/p38 imbalance and contributing to high ERK activity throughout the entire period of
tumor growth. In contrast, in the small nodules formed by dormant cells the level of ERK activity was dramatically reduced, while p38 activity remained high.
Strong activation of ERK was evident in metastatic sites, while p38 activation was silenced in this anatomical location as well. These results show that it is
possible to directly measure cancer cell response to microenvironment using this reporter system and that only proliferation competent cells have the ability to
rapidly adapt ERK and p38 signaling for proliferative success. This approach allows isolation and further characterization of metastatic cells with specific signaling
signatures indicative of their phenotypes.
|
|
Fig3. (A) T- (tumorigenic) and D-Elk (dormant) cells were grown in culture and inoculated on CAMs (5x105 cells/CAM). Cells in culture (a and e), 24 hrs on CAMs
(b and f) and 4 days on CAMs (d and g). Tumor nodules captured in GFP channel (a, b, e, f) at x400. A x200 image of the same tumor areas are shown in d and g.
Scale bars in a, b, e and f= 40 µm; 160 µm in d and h. (B) GFP positive T-Elk cells in a nude mice lung at the time of primary tumor removal; scale bar 40 µm.
(C) FACS analysis (mean GFP fluorescence) of T-Elk and D-Elk cells, grown in culture with or without the Mek inhibitor PD98059 (40 µM). The graph shows mean+SE.
|
Stress-signaling through p38-SAPK: The ability of cells or organisms to sense environmental (or micro-environmental stress in the case of cells) stress allows them to adapt to
conditions that may not favor proliferation or in the case of organisms, development and the production of offspring. Thus, activation of stress signals in tumor cells may allow them to adapt to new
microenvironments (i.e. during dissemination) or stress insults such as those they encounter during chemotherapy. As noted above, activation of p38 results in inhibition of ERK and under such
conditions G0/G1 arrest and dormancy are induced. These results pointed to signaling activated by the stress kinase p38 as important for the induction of dormancy. We are using proteomic and
genomic analysis to identify the mechanisms by which p38 regulates induction of dormancy and survival of dormant cells (Fig4). By proteomic comparison of HEp3 cells with high (dormant) vs. low p38 (tumorigenic) activity we discovered that p38 signaling is linked to the activation of endoplasmic reticulum (ER) stress signaling. These stress signals are usually activated during a response known as the integrated stress response (ISR) and it is known to induce G0/G1 arrest following ER-stress. Further, our proteomic search revealed that, in dormant cells, p38 signaling alters the shuttling of heterogeneous nuclear ribonucleoproteins (hnRNPs), which regulate post-transcriptional gene expression of mRNAs. Our genomic analysis also revealed that activation of p38 in dormant cells induces tumor suppressors known to regulate G1->S or G2->M progression and also survival factors indicating that p38 signaling appears to activate a program that concomitantly coordinates growth arrest
|
|
Fig4.Hypothetical model depicting how stress signals that result in p38 activation, can suppress ERK signaling. Further, active p38 appears
to activate an ER-stress response that coordinates growth arrest, through the activation of molecules such as PERK and survival through for example inhibitors of
apoptosis (IAPs). These pathways fulfill the two requirements of dormancy, growth arrest and robust survival.
|
and robust survival. We are studying the mechanisms by which p38 stress-activated signaling favors dormancy and survival. We are also studying how these mechanisms may favor resistance of dormant cells to chemotherapy (Fig4). Cancer genomics: We are also testing in collaboration with Dr. Douglas Conklin’s laboratory (
http://www.albany.edu/cancergenomics/faculty/dconklin/dconklin.html)
the use of a library of shRNAs to discover the function of a third of the genes present in the human genome to the malignant behavior of various malignant human
cancers and their role in the dormant behavior of tumor cells.
Tumor cells that disseminate to distant organs can use adhesion-dependent signaling to feed information from their new surroundings. Signals relayed by integrins from the
extracellular matrix can determine if cells proliferate, die or stop growing. Our lab is studying how integrin signaling may determine the fate of tumor cells that disseminate to
different organs. Specifically we are trying to understand how integrin-activated signaling through stress activated pathways, such as JNK or p38 induce growth arrest and/or
apoptosis signals (Fig5). We are also testing whether integrin mediated inhibition of mitogenic pathways may contribute to the induction of tumor cell dormancy in vivo.
 |
|
Fig5. Model depicting how integrins might relay signals that result in growth inhibition of tumor cells.
Activation of p38 and/or JNK in response to specific ECM (fibrillar collagen)-integrin cues regulates a growth arrest and
survival gene expression program. Concomitantly, the same or alternate ECM-integrin cues can, through p38 or other mechanisms,
suppress growth- promoting signals. Together these signals result in a G0/G1 arrest and induction of dormancy..
|
Startup Funds from the State University of New York at Albany
Samuel Waxman Cancer Research Foundation
|
