CMRC Children's Memorial Research Center
Research Projects

Description of Funded Research Projects Ongoing
in the Hendrix Laboratory



During embryonic development, the formation of primary vascular networks occurs by vasculogenesis -- the in situ differentiation of mesodermal progenitor cells to endothelial cells that organize into a primitive network. The subsequent remodeling of the vasculogenic network into a functionally efficient vasculature occurs through angiogenesis -- the sprouting of new capillaries from a preexisting network. Our laboratory and collaborators have introduced the term “vasculogenic mimicry” to describe the unique ability of aggressive melanoma tumor cells to form tubular structures and patterned networks in 3-D culture, which “mimics” the pattern of embryonic vasculogenic networks and recapitulates the patterned networks seen in patients’ aggressive tumors -- correlating with poor prognosis (1). The molecular profile of these aggressive tumor cells suggests that they have a deregulated genotype, capable of expressing multiple molecular phenotypes simultaneously, particularly those associated with endothelial cells and their precursors. In addition, our preliminary studies indicate that: 1) aggressive melanoma cells express VE-cadherin (exclusively expressed by endothelial cells) -- critical in the formation of vasculogenic networks; 2) the aggressive tumor cells produce an extracellular matrix (ECM) that induces poorly aggressive melanoma cells to form vasculogenic networks; and 3) the aggressive melanoma cells participate in the revascularization of an ischemic limb model, thus suggesting a stem cell plasticity potential for these cells, resembling endothelial cells. The proposed studies advance observations made during the current funding period regarding the embryonic-like phenotype of aggressive melanoma cells, and focus on the development of new diagnostic approaches for the detection of tumor cells masquerading as endothelial cells. The data generated from these novel studies will provide new molecular markers for clinical diagnosis in addition to new concepts regarding the transendothelial differentiation of aggressive melanoma tumor cells and their stem cell plasticity.


Despite significant efforts to develop valid predictors of breast cancer metastatic potential, there remains a critical gap in our understanding of the molecular function(s) of the genes involved in the maintenance of the normal human mammary epithelial cell (HMEC) phenotype, many of which are altered during the onset and progression of breast cancer. This competitive renewal applicatioin advances our knowledge of the biological function(s) of maspin (mammary serpin; a putative tumor suppressor gene) and its molecular interactions with a newly identified binding partner, IRF6 (Interferon Regulatory Factor 6; a transcription factor) -- and focuses on determining their independent and collective mechanistic role(s) during normal mammary gland acini development and the consequence(s) of their loss during breast cancer progression. Special emphasis will be placed on determining the role(s) of IRF6/maspin on cellular interactions with 3-D matrices to decipher the epigenetic effects of the remodeled microenvironment on cellular phenotype and behavior. The translational value of these studies rests in the development of new strategies to re-express tumor suppressor in breast cancer cells that result in the neutralization of the differentiation and promigratory signals in the microenvironment.


The long range goal of this project is to develop intervention strategies based on newly discovered biological mechanisms responsible for the invasive dissemination of metastatic uveal melanoma. To accomplish this goal, a multifaceted approach will be followed for on determining the biological relevance of novel marker proteins contributing to the metastatic phenotype of uveal melanoma. Melanoma cells typically express vimentin intermediate filaments (IFs); whereas, epithelial cells typically express keratin IFs. Cells that coexpress both IFs are considered to be "interconverted" -- both mesenchymal and epithelial phenotypes. Although the biological functions of IFs have remained enigmatic, there is substantial support to suggest that the significance of vimentin/keratin coexpression is linked with poor patient outcome. The epithelioid morphology in uveal melanoma is perhaps explained by the coexpression of keratin and vimentin. Preliminary data demonstrate that uveal melanoma cells which coexpress vimentin and keratin IFs exhibit an interconverted phenotype and are 6-fold more invasive through extracellular matrices and induce endothelial angiogenic structures, compared with cells expressing vimentin; are predominantly found in the liver metastases of patients with uveal melanoma; and overexpress the c-met proto-oncogene -- the receptor for HGF/SF (hepatocyte growth factor/scatter factor, which is a potent motogenic, mitogenic and morphogenic factor). Based on these novel and exciting observations, the proposed study will test the overall hypothesis that coexpression of vimentin and keratin IFs by uveal melanoma cells contributes to a highly invasive and metastatic interconverted phenotype, which is regulated by c-met and HGF/SF.


Background and Significance: The alarming increased incidence in prostatic and ovarian carcinoma must be addressed from the perspectives of improving diagnostic technology, better therapeutic management of metastatic disease, and, ultimately, the development of prevention strategies. Although we are unsure of the etiology of these diseases, we do know that there is a higher incidence with age and that there is an association with the consumption of Western diets (which affects fat intake and hormonal balance). Furthermore, there is an epigenetic silencing of naturally occurring tumor suppressor genes in these cancers (probably due to aberrant cytosine methylation and heterochromatinization) -- which causes genes such as E-cadherin to lose its expression in normal prostatic and ovarian epithelium and induce the cells to produce increased levels of MMPs (proteases) and invade and metastasize.
Experimental approach: Using sophisticated molecular biology techniques, we plan to analyze the genetic profiles of aggressive vs. nonaggressive human prostatic and ovarian carcinoma to provide new gene targets that are responsible for the cancerous properties of these two diseases. These specifically identified genes will be rigorously tested for their ability to induce prostatic and ovarian carcinoma in appropriate animal models, followed by an attempt to inhibit this induction using these same models. Simultaneously, we will analyze patient tumors for the expression of these key genes to determine whether they are associated with disease progression -- and whether they could be utilized as diagnostic markers to assess the aggressiveness of prostatic and ovarian carcinoma. If the information we generate from all facets of this study indicate that specific genes such as MMPs and E-cadherin are critical molecular determinants of these cancers, then we will proceed with establishing innovative clinical trials focused on inhibiting disease progression and preventing disease occurrence.
Benefit to society: Based on the scientific data expected from our investigation, biomedical research would benefit from the discovery of new genes associated with prostatic and ovarian carcinoma, the elucidation of key molecular mechanisms that underlie these cancers, and the knowledge to develop new and improved diagnostic methods to detect the onset of disease at the earliest possible stages. Most importantly, we will gain knowledge related to the prevention of these devastating cancers, which will benefit posterity.


Recently, our laboratory has described a new phenomenon, tumor vasculogenesis, whereby tumor cells themselves (in the absence of endothelial cells), form vascular channels and tubular networks which facilitate tumor perfusion independent of classical angiogenesis. The original discovery was made in human uveal and cutaneous melanoma, both in patient tumors and in 3-dimensional in vitro models using patient tumor derived cell lines (1). However, our preliminary studies utilizing prostate tumors and neoplastic prostate cell lines strongly support the concept that "vasculogenic mimicry" is exhibited by aggressive, but not nonaggressive, prostatic neoplasms, as well. It is, therefore, the overall objective of the proposed research plan to determine the key molecular mechanisms underlying this phenomenon in tumors of the prostate. The current proposal advances our previous observations and specifically addresses fundamental molecular mechanisms underlying vasculogenic mimicry by aggressive prostate cancer cells. Based on our preliminary studies in prostate cancer (in both rat and human models), we have discovered that: 1) Invasive prostate cancer cells form tubular networks with vascular channels and conduct dyes reminiscent of a microvascular circulation; 2) in heterogeneous prostate cancers, subpopulations of epithelial and fibroblast-like cells interact cooperatively to form the tubular networks and vascular channels, which appear to anastomose with endothelial lined vessels; 3) the prostate cells undergoing vasculogenic mimicry express key endothelial markers; and 4) one of the critical mechanisms involved in vasculogenic mimicry is the expression of specific matrix metalloproteinases (MMPs), based on our observations that downregulation of MMP activity abrogates prostatic vasculogenic mimicry. The data generated from these novel studies will provide new molecular markers for diagnosis in addition to new concepts in prostate cancer research with respect to metastasis and clinical intervention strategies. 

Role of Maspin in Human Placental Development (K08 HD42666 to Anuja Dokras)

The human placenta displays highly regulated invasive activity and an exponential growth potential, both of which are temporally regulated. Cytotrophoblasts invade the endometrium, myometrium and vascular spaces primarily in the first and second trimester and to a lesser extent in the third trimester of pregnancy. The process of invasion demonstrated by the extravillous cytotrophoblasts has been shown to be regulated by matrix metalloproteases (MMPs), and associated with changes in the expression of integrins, and insulin-like growth factor I. Despite this extensive spatial invasion, the risk of developing invasive mole and choriocarcinoma is low (1 in 20,000-40,000 pregnancies). On the other hand, decreased invasion by cytotrophoblasts can result in clinical syndromes such as preeclampsia, and is also associated with pregnancies complicated by intra-uterine growth restriction (IUGR). Identification of factors that play a role in modifying the invasive ability of cytotrophoblasts will contribute to a better understanding of this unique developmental process.

Maspin (mammary serine protease inhibitor) is a tumor suppressor gene with inhibitory actions on motility, invasion, metastasis and angiogenesis in human breast and prostate cancer cells. Loss of maspin expression in these cancers correlates with tumor invasiveness and breast cancer recurrence. Recently, we have found that maspin is differentially expressed in the human placenta with maximal expression in the third trimester, which is the period of least cytotrophoblast invasion. There is no other information in the literature on the role of maspin in the human placenta or the female reproductive tract. Our data suggest that maspin may also play a role in the regulation of cell invasion in the human placenta. The down-regulation of maspin expression may be critical at the time of implantation and early placental development, whereas up-regulation of maspin may limit this process at the end of gestation. We hypothesize that the coordinated differential expression of maspin during pregnancy coincides with specific cytotrophoblast functions, specifically, endometrial invasion. Further, the expression of maspin is regulated, at least in part, at the transcriptional level by conditions such as hypoxia during precise periods of placental development.

The long-term objectives of this project are two-fold: first, to understand the significance and mechanism(s) of action of tumor suppressor genes such as maspin in human placental development and second, to identify pathological processes during gestation arising from alterations of maspin expression. The latter could have profound implications on therapeutic strategies for clinical conditions associated with decreased cytotrophoblast invasion such as preeclampsia and intra-uterine growth restriction (IUGR).