Dai Fukumura, MD, PhD
Associate Professor, Department of Radiation Oncology, Harvard Medical School
Deputy Director, Edwin L. Steele Laboratory, Radiation Oncology, Massachusetts General Hospital
Massachusetts General Hospital
100 Blossom St
Boston, MA, 02114
Massachusetts General Hospital
100 Blossom St
Boston, MA, 02114
DF/HCC Program AffiliationGastrointestinal Malignancies
Angiogenesis, Invasion and Metastasis
Research AbstractThe long-term goal of my research is to uncover the fundamental nature of vascular biology in both physiological and pathophysiological settings, and to utilize this knowledge for the treatment of vascular disorders, metabolic and malignant diseases, and tissue engineering. To this end we have been developing and utilizing state of the art imaging techniques and animal models that led us to uncover novel insights in angiogenesis, lymphangiogenesis, vascular function, microenvironment and treatment response as summarized below.
Development of imaging techniques and in vivo models
In order to dissect molecular, cellular, morphological and functional insights necessary for the study of vascular biology, host-tumor interaction and its influence on the growth, metastasis, treatments of tumors, developments of novel imaging techniques and sophisticated animal models, which faithfully represent clinical behavior of tumors, are absolutely essential. To this end we are continuously establishing and utilizing state of the art intravital microscopy techniques (Nature Reviews Cancer 2: 266, 2002) including in vivo fluorescent protein gene reporter systems (Cell 94: 715, 1998; Cancer Res 61: 6248, 2001; Cancer Res 64: 5920, 2004; PNAS 108: 18784, 2011), multiphoton laser-scanning microscopy (MPLSM) (Nature Medicine 7: 864, 2001) which allows us to observe deep inside the tumor with high spatial resolution, video-rate MPLSM (IntraVital 1: 60, 2012), modification and application of quantum dots (Qds) (Nature Medicine 11: 678 2005), optical frequency domain imaging (OFDI) (Nature Med 15: 1219, 2009) which is extremely robust and tracer free imaging technique, and provides complementary data to MPLSM, innovative in vivo endomicroscopy (Nature Methods 7:303, 2010), and multi-parametric functional analysis techniques (Nature Methods 7: 655, 2010), and various sizes and configuration of Qd-base nanoparticles (Biophys J 99: 1342, 2010; Angew Chem Int Ed Engl 49: 8649, 2010; PNAS 108: 2426-31, 2011; Angew Chem Int Ed Engl 50:11417, 2011). We have also developed a number of new animal models including intravital microscopy models for liver tumors (American Journal of Pathology 151: 679, 1997) and breast cancers (Clin Cancer Res 8: 1008, 2002), abdominal window model for pancreatic cancers (Laboratory Investigation 81:1439, 2001), tissue-engineered blood vessel model (Nature 428: 138, 2004), spontaneous colorectal cancer model & micro-endoscopy (Nature Methods 7: 303, 2010). In addition, we have established quantitative blood assay for viable tumor burden using secretive Gaussia luciferase (Gluc) (PLoS One 4: e8313, 2009).
Role of NO in tumor angiogenesis, vascular function and anti-tumor therapy.
Nitric oxide (NO) is a highly reactive mediator with a variety of physiological and pathological functions as we summarized in (Nature Reviews Cancer 6: 521, 2006). We revealed that NO increases and/or maintains tumor blood flow, decreases leukocyte-endothelial interactions, and increases vascular permeability and thus, may facilitate tumor growth (Am J Pathol 150: 713, 1997). Furthermore, we demonstrated that NO mediates angiogenesis (PNAS 98: 2604, 2001) and vessel maturation (JCI 115: 1816, 2005) predominantly through endothelial NO synthase. We also found that NO mediates lymph-angiogenesis and metastasis (Cancer Res 69: 2801, 2009) as well as function of lymphatic vessels (Circulation Research 95: 204, 2004). We recently uncovered that non-vascular sources of NO derived from tumor cells or inflammatory cells disrupts perivascular NO gradient and impairs blood vessel maturation and function (Nature Med 14: 255, 2008) as well as lymphatic vessel function (PNAS 108: 18784, 2011). Restoration of perivascular gradient of NO recovers lymphatic vessel contraction (PNAS 108: 18784, 2011), and normalizes structure and function of tumor vessels improving tissue oxygenation and response to radiation therapy (Nature Med 14: 255, 2008).
Role of tumor-host interactions in angiogenesis, tumor growth, metastasis and treatments.
Using transgenic mice harboring the green fluorescent protein (GFP) gene driven by vascular endothelial growth factor (VEGF) promoter we found that VEGF promoter of non-transformed stromal cells is strongly activated by the tumor microenvironment (Cell 94: 715, 1998). Using tumor cells carrying the same gene construct we found, for the first time, that hypoxia and low pH independently upregulate VEGF in vivo (Cancer Res 61: 6248, 2001; J Biol Chem 277: 11368, 2002). Using VEGF-/- and wild type ES cell derived tumors we found that the host cells contribute approximately half of total VEGF production in this model (Cancer Res 60: 6248, 2000). MPLSM revealed that VEGF expressing stromal cells are closely associated with angiogenic vessel in the tumor (Nature Med 7: 864, 2001). The association of VEGF-expressing stromal cells spatially correlates with the extravasation of nanoparticles (Nature Med 11: 678, 2005). Furthermore, various anti-tumor treatments result in increased expression of host stromal cell VEGF and thus, may contribute to treatment resistance (Nature 416: 279, 2002; PNAS 2012 Oct 15. [Epub ahead of print]). In fact, judicious blockade of VEGF signaling can transiently normalize tumor vasculature and potentiate radiation therapy (Cancer Cell 6: 553, 2004; PNAS 108: 1799, 2011). Anti-VEGF treatments prolong survival of brain tumor bearing animals by normalizing the vasculature – reducing edema – (J Clin Oncol 27: 2542, 2009). However, ectopic expression of Angiopietin-2 compromises this benefit (Clin Cancer Res 16: 3618, 2010). On the other hand, platelet derived growth factor-D normalizes tumor vessels and improves delivery and efficacy of chemotherapeutics (Clin Cancer Res 17: 3638, 2011). Finally, anti-VEGF treatment can restore immune microenvironment in tumors and potentiate vaccine therapy (PNAS 109:17561-6, 2012).
We have been further dissecting mechanisms of tumor escape from anti-VEGF treatment focusing on inflammatory cells and pathways that can aggravate tumor growth, metastasis and response to treatment. Our recent data indicate that stromal cells in the primary tumor travel with tumor cells and facilitate survival and growth of metastatic tumors (PNAS 107: 21677, 2010). However, VEGFR1+ bone marrow derived cells (BMDCs) are not required for spontaneous metastasis (Nature 461: E4, 2009; PLoS One 4: e6525, 2009). We found that CXCR4 promotes metastasis via Gr-1+ BMDC recruitment (PNAS 108: 302, 2011). Finally, we found that metastatic tumors induce focal hyper-permeability in the lungs, leading to regional accumulation of inflammatory cytokines, and creating preferential sites of metastatic “soil” (PNAS 108: 3725, 2011).
Probing tumor microenvironment using nanotechnology
We have been studying the tumor microenvironment and transport properties of nanoparticles using Qd-based probes (Nature Medicine 11: 678, 2005). Using the size series system (~10, ~60, ~120 nm), we have measured the transvascular flux and plasma clearance of the nanoparticles. Although all of these nanoparticles extravasated, large nanoparticles (~120 nm) stayed within 10 µm of the tumor vessel wall (Angew Chem Int Ed Engl 49: 8649, 2010). We also found superior transvascular transport properties of rod-shaped nanoparticles than spherical shape nanoparticles with the same hydrodynamic size (Angew Chem Int Ed Engl 50: 11417, 2011). Furthermore, we determined the effect of charge on nanoparticle transport and found that cationic and neutral charge is the best for transvascular and interstitial transport, respectively (Biophys J 99: 1342, 2010; Ann Biomed Eng [Epub] 2012). These findings let to an approach to exploit tumor microenvironment. Nanoparticle drug delivery systems are based on enhanced permeability and retention (EPR) effect of tumor vasculature and can improve therapeutic index. However, these particles suffer heterogeneous distribution and lack of tissue penetration after extravasation (Nature Nanotechnol 7: 383, 2012). In order to circumvent this critical shortcoming, we have developed multistep nanoparticles that shrink upon exposure to the tumor microenvironment in order to facilitate delivery and efficacy of therapeutics to tumors. To this end we have recently demonstrated a proof of principle that gelatin nanoparticles are digested and release quantum dots in MMP-2 expressing tumors (PNAS 108: 2426, 2011).
Role of obesity in angiogenesis, tumor growth and treatments.
First, we established in vivo system to investigate blood vessel formation during adipogenesis. Using genetic inhibition of PPARγ and pharmacological inhibition of VEGFR2 signaling we found provocative reciprocal regulation of adipogenesis and angiogenesis, suggesting a novel strategy to treat obesity related diseases including cancer (Circulation Res 93: e88, 2003; PLoS One 4: e4974, 2009). We then established a physiologically based mathematical model and found that leptin pathway plays a key role in maintenance of body mass and its disruption destroys the body weight balance (Cell Metabolism 9: 52, 2009). We are currently studying the underling mechanisms of obesity-induced aggravation of breast cancer through both preclinical studies and clinical trials of breast cancer patients.
Engineering blood vessels.
A major limitation of tissue engineering is the lack of functional blood and lymph vessels. First, we established a model to monitor tissue engineered blood vessels in vivo using MPLSM. We found that mesenchymal precursor 10T1/2 cells accelerate remodeling of 3-D endothelial cell structure to functional blood vessels, differentiate into peri-vascular cells, and stabilize engineered vessel network for up to a year (Nature 428:138, 2004). We then, showed that human ES cell, cord blood and peripheral blood -derived endothelial cells form functional blood vessels in vivo using the tissue engineered blood vessel model (Nature Biotech 25: 317, 2007; Blood 111: 1302, 2008). Establishment of functional and stable blood vessels was advanced by the success of transplantation of human bone marrow derived mesenchymal stem cells that served as perivascular precursor cells (Blood 111: 4551, 2008). On the other hand, it was paradoxically inhibited by endothelial overexpression of PDGF-B (Am J Pathol 175: 294, 2009). Detail observation of vessel anastomosis in these tissue-engineered blood vessels revealed a novel mechanism – wrapping-and-tapping of host vessels (Blood 118: 4740, 2011). More recently, we have established robust protocols deriving endothelial cells and mesenchymal precursor cells from induced pluripotent stem (iPS) cells and successfully generated blood vessels in vivo from these iPS-derived cells.
Significant portion of my time has also been devoted to teaching tumor angiogenesis and microenviornment as well as various experimental techniques through daily supervision of research fellows and graduate students. My trainee are very successful as evidenced by their outstanding publications in Nature, Nature Biotechnology, Nature Medicine, Nature Methods, Journal of Clinical Investigation, Blood and Journal of Clinical Oncology as well as their awards from NIH, DOD, Howard Hughes Medical Institute, American Association for Cancer Research, Institute for Cancer Research, Susan Komen Foundation, Massachusetts Biomedical Research Corporation, Deutsche Forschungsgemeinschaft, La Ligue Nationnale Contre le Cancer, and Japanese Ministry of Health and Welfare. I also provide multiple lectures both in local and international courses and workshops.