| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Surgery, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA
1Correspondence: Department of Surgery, University of Cincinnati College of Medicine, 231 Albert Sabin Way, Cincinnati, OH 45267-0558, USA. E-mail: alex.lentsch{at}uc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: cancer • angiogenesis • African-American • endothelial cell
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. There is abundant evidence to suggest that erythroid DARC functions to bind and clear chemokines from sites of overproduction, such as inflammatory foci (2
, 3)
. In addition to its chemokine binding capacity, erythroid DARC functions as a receptor for Plasmodium vivax infection (1)
. The function of the DARC expressed on endothelial cells is not well understood. It has been suggested that endothelial cell DARC facilitates the transport of chemokines from tissue to the vascular lumen for presentation (4)
. However, the binding of chemokines by endothelial DARC does not transduce any signal nor does it result in altered endothelial cell function (4)
.
The DARC binds to a subset of CXC chemokines containing the amino-terminal sequence Glu-Leu-Arg, also known as the ELR motif (5)
. Site-directed mutagenesis studies have shown that the ELR motif confers angiogenic properties to CXC chemokines, primarily via endothelial cell chemotaxis (6
, 7)
. A variety of ELR+ CXC chemokines are produced by prostate cancer cell lines (8
, 9)
, and increased levels of ELR+ CXC chemokines have been detected in tumor and sera of patients with prostate cancer, but not prostate hyperplasia (10
, 11)
.
The vast majority of the population of African descent lack expression of the DARC on erythrocytes as a natural selection against malarial infection. As such, individuals who lack erythroid expression of the DARC, as occurs in > 95% of African men in endemic regions and in 70% of African-American men, may have a higher concentration of angiogenic CXC chemokines in the microenvironment of a developing prostate tumor. To test this hypothesis, we examined the involvement of the DARC in prostate tumor growth in the transgenic adenocarcinoma of the mouse prostate (TRAMP) model (12)
.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Conditioned media from PC-3 cells cultured for 48 h was incubated with nothing or 4.8 x 107 erythrocytes from either DARC+/+ or DARC–/– mice for 1 h at 37°C. Erythrocytes were pelleted by centrifugation. Concentrations of the human chemokines, CXCL1, CXCL8, CXCL10, CXCL11, as well as vascular endothelial cell growth factor (VEGF), were determined by ELISA (all reagents from R&D Systems, Minneapolis, MN, USA). Alternatively, postincubation conditioned media was assessed for chemotactic activity toward human dermal microvascular endothelial cells (HMEC-1; Center for Disease Control, Atlanta, GA, USA) using methods described elsewhere (6)
.
Transgenic mouse model of prostate cancer
The TRAMP mouse model of prostate cancer was used in this study (12)
. DARC+/+ and DARC–/– mice were generated as described elsewhere (3)
. TRAMP mice were crossed with DARC+/+ or DARC–/– mice to obtain TRAMP mice expressing the DARC (+/+) or TRAMP mice lacking DARC expression (DARC–/–). Prostate tumors from mice were harvested at 20, 25, and 30 wk of age. This project was approved by the University of Cincinnati Animal Care and Use Committee and conforms to the National Institutes of Health guidelines.
Tumor analysis
Tumor volume was calculated by measuring three radii and applying to the formula for spheroid volume (4/3
r3). Tumor mass was determined by weighing the tumor after excision. Intra-tumor concentrations of macrophage inflammatory protein-2 (MIP-2), keratinocyte-derived chemokine (KC), and VEGF were determined by ELISA (R&D Systems).
Real-time RT-PCR
One microgram of total RNA from prostate tumor tissue was reverse transcribed to cDNA using a SuperScriptTM First-Strand Synthesis System (Invitrogen, San Diego, CA, USA) according to the manufacturer’s instructions. Two microliters of cDNA was amplified by PCR using primers for MIP-2, KC, and ß-actin. PCR was performed using a real-time PCR system (Mx3000PTM, Stratagene, La Jolla, CA, USA) and the following conditions: predenaturation at 95°C for 10 min; denaturation at 95°C for 30 s; annealing at 58°C for 1 min; extension at 72°C for 45 s. After repeating for 40 cycles, redenaturaton was performed at 95°C for 1 min. The optimal concentration of primers was determined by a performance curve, and PCR efficiency and the amplified gene quantity were determined by two different standard curves.
Tumor vessel density
Deparaffinized tumor sections were digested for antigen retrieval with proteinase K, washed with PBS, and blocked with 2% goat serum in PBS. Sections were incubated with an antibody against endothelial cells (rabbit anti-VWF, Dako, Glostrup, Denmark), washed, then incubated with a biotinylated antibody (goat anti-rabbit, Vector Laboratories, Burlingame, CA, USA). After washing, sections were incubated with avidin-phosphatase and Vector red substrate buffer according to the manufacturer's instructions (Vector Laboratories). Sections were examined by fluorescence microscopy, digitally photographed, and quantitated using a histogram analysis of the red channel fluorescence.
Statistical analyses
Data are expressed as mean ± SE. Data were analyzed with a 1-way ANOVA and a subsequent Student’s t test. Differences were considered significant when P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Conditioned media from PC-3 cells cultured for 48 h were analyzed for the angiogenic chemokines CXCL1 and CXCL8 before and after incubation with erythrocytes from DARC+/+ or DARC–/– mice. To control for nonspecific binding of other mediators that may affect the angiogenic potential of the cultured media, we also measured the levels of the nonangiogenic (ELR–) chemokines, CXCL10 and CXCL11, as well as VEGF before and after incubation with erythrocytes from DARC+/+ or DARC–/– mice. The presence of DARC+/+ erythrocytes significantly reduced the levels of angiogenic chemokines (Fig. 1
A), but not nonangiogenic (ELR–) chemokines (Fig. 1B
) or VEGF (Fig. 1C
), in conditioned media.
|
We next determined whether the clearance of angiogenic chemokines by DARC+/+ erythrocytes altered the chemotactic activity of these conditioned media toward endothelial cells. We found that conditioned media from cultured PC-3 cells resulted in significant endothelial cell chemotaxis (Fig. 1D
). However, when the same media was incubated with DARC+/+ erythrocytes, the amount of endothelial cell chemotaxis was significantly reduced (Fig. 1D
). When the media was incubated with DARC–/– erythrocytes, the amount of endothelial cell chemotaxis was identical to that of media alone (Fig. 1D
). These data demonstrate that the DARC binds and clears angiogenic chemokines produced by prostate tumor cells and reduces the ability of prostate tumor cell secretions to induce endothelial cell chemotaxis. Thus, the DARC may function in vivo to reduce the angiogenic potential within a prostate tumor.
Gene deletion of the DARC results in augmented prostate tumor growth
To explore the potential contributions of the DARC to the development of prostate cancer in vivo, we used a novel model system in which DARC–/– mice were crossed with TRAMP mice. TRAMP mice spontaneously develop prostate adenocarcinoma within 20–30 wk that is histologically similar to the human disease (12)
. To examine the role of the DARC in the onset and development of prostate cancer, we crossed TRAMP mice with DARC+/+ mice and DARC–/– mice. These crosses produced TRAMP mice that express the wild-type DARC gene (TRAMP/DARC+/+) and those that are nullizygous for the DARC (TRAMP/DARC–/–). At 20 wk of age there was no statistical difference in the size or mass of prostate tumors in TRAMP/DARC+/+ vs. TRAMP/DARC–/– mice (Fig. 2
A). However, at 25 and 30 wk of age, tumor volume was > 4-fold greater and tumor mass was > 2-fold greater in the TRAMP/DARC–/– mice compared with TRAMP/DARC+/+ mice (Fig. 2A
, B).
|
Prostate tumors in DARC knockout mice have higher concentrations of angiogenic CXC chemokines and greater vascular density
To determine whether the DARC may be modulating prostate tumor growth through the binding and clearing of angiogenic chemokines from the tumor microenvironment, we measured tumor concentrations of the murine angiogenic chemokines MIP-2 and KC, both of which have high sequence homology with the human growth-related oncogenes (CXCL1-CXCL3) and functional homology with human CXCL8 (13)
. Tumor concentrations of the nonchemokine angiogenic factor VEGF, were also determined. The concentration of these factors in normal prostate tissues from 25-wk-old wild-type (TRAMP-negative) C57BL/6 mice was very low (Fig. 3
A). Prostate tumor concentrations of MIP-2 and KC in TRAMP/DARC+/+ mice were 
10-fold higher than normal (Fig. 3A
). Levels of VEGF were 3-fold higher in prostate tumors from TRAMP/DARC+/+ mice compared with normal prostate from normal wild-type mice. In tumors from TRAMP/ DARC–/– mice, MIP-2 and KC levels were significantly greater than TRAMP/DARC+/+ mice. Intra-tumor concentrations of VEGF did not differ between TRAMP/DARC+/+ and TRAMP/DARC–/– mice (Fig. 3A
).
|
To confirm that the increases observed in tumor MIP-2 and KC protein levels were not due to alterations in gene expression, we examined tumor expression of MIP-2 and KC mRNA by real-time RT-PCR. As Fig. 3B
shows, there were no differences in the expression of MIP-2 or KC mRNA in tumors from TRAMP/DARC+/+ or TRAMP/DARC–/– mice throughout the development of tumor formation.
To determine whether the increased concentrations of angiogenic chemokines found in the tumors of TRAMP/DARC–/– mice were associated with increased angiogenesis, we analyzed tumor vessel density by staining for vascular endothelial cells. Tumors from TRAMP/DARC+/+ and TRAMP/DARC–/– mice had abundant vessel staining. However, there appeared to be more vessel staining in tumors from TRAMP/DARC–/– mice than in TRAMP/DARC+/+ mice (Fig. 4
A). Acquired images of tumor vessel staining were assessed using a histogram analysis of the red channel. This analysis found that tumors of TRAMP/DARC–/– mice had significantly greater vessel density than TRAMP/DARC+/+ mice (Fig. 4B
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. There is ongoing debate as to whether the increased risk and mortality to prostate cancer in African-American men is due to biological, environmental, or socio-economic factors. To date, there has been little credible proof that any of these are significant contributing factors. The present studies provide strong evidence of a biological mechanism for the increased mortality to prostate cancer in men of African descent.
There is growing clinical data indicating that angiogenic (ELR+) CXC chemokines contribute to the pathogenesis of prostate cancer. Immunohistochemical studies of human prostate tissues have shown that prostate tumors express abundant amounts of CXCL8, whereas normal prostate and benign prostatic hyperplasia tissues express virtually none (10)
. Serum levels of CXCL8 are not increased in patients with benign prostatic hyerplasia, but in patients with prostate cancer, serum CXCL8 correlate with increasing stage of prostatic disease (11)
. Furthermore, prostate cancer cell lines produce abundant amounts of angiogenic CXC chemokines (8
, 9)
, implicating the use of these chemokines by prostate cancer cells to promote tumor growth.
We propose that the greater mortality to prostate cancer in men of African descent is due in part to the genetic mutation that prevents expression of the DARC on erythrocytes that evolved long ago as a natural selection against malaria infection (15)
. In vitro studies have demonstrated that erythrocyte DARC binds to multiple members of the chemokine family (5)
. In the current studies we found that erythroid DARC bound to ELR+ chemokines produced by prostate cancer cells in vitro and reduced the endothelial chemotactic activity of the conditioned media from these cell cultures, suggesting that erythroid DARC has the capacity to reduce the angiogenic potential of prostate tumor secretions. Our data also show that erythroid DARC binds only to ELR+ chemokines, not ELR– chemokines or the cytokine VEGF.
Our in vitro studies set the stage for experiments with DARC–/– mice to determine whether the DARC plays a role in the growth of prostate tumors in vivo. Using the TRAMP model of prostate cancer in mice, we found no differences in the onset of prostate cancer in DARC+/+ vs. DARC–/– mice. This is not surprising as chemokines have not been linked to initiation of cancer (i.e., transformation). Furthermore, in vitro studies have demonstrated that CXC chemokines do not have proliferative effects on prostate tumor cells (8)
. The data suggest that the DARC has no role in the initiation of prostatic disease but has a profound effect on subsequent tumor development. Our data showing dramatic increases in prostate tumor growth in DARC–/– mice provide strong evidence that the DARC has an important regulatory function in this process. Our findings of increased concentrations of ELR+ CXC chemokines (but not VEGF) and increased vascular density in prostate tumors of DARC–/– mice support the concept that the increased tumor growth in DARC–/– mice is directly related to an inability of these mice to clear angiogenic chemokines from the tumor microenvironment.
A potential limitation of the present model, and its comparison or extrapolation to men of African descent, is the total lack of DARC expression in DARC–/– mice. The human mutation that protects against malarial infection results only in deletion of the erythoid form and leaves intact the endothelial form (15)
. The relative roles of erythroid vs. endothelial DARC in regulating chemokine-mediated events is not well understood. Some studies, including our present data, support the concept of erythroid DARC as a "sink" for chemokines. The function of endothelial DARC appears to be more complex. Lee et al. have shown that endothelial DARC may function to transport chemokines across the endothelial barrier to maintain tissue chemokine gradients (4)
. Other studies have shown that endothelial DARC localizes to caveolae and that chemokines bound by endothelial cells are localized to caveolae, internalized, and subsequently transported to the luminal surface (16
, 17)
. It has been proposed that in addition to chemokine transport, endothelial DARC functions to "present" chemokines to other receptors. However, there are data that refute this possibility and suggest that both endothelial and erythroid DARC quench the activities of chemokines in vivo. Two reports show that chemokines bound to DARC are not accessible to other chemokine receptors, indicating that the DARC does not present or transfer bound chemokines to signaling receptors (5
, 18)
. Other reports show that DARC transgenically overexpressed in endothelial cells or tumor cells reduces angiogenesis, implicating a scavenging effect of the DARC on angiogenic CXC chemokines (19
, 20)
.
In summary, our data provide strong evidence that the DARC is a key modulator of the progression of prostate cancer by clearing angiogenic chemokines from the tumor microenvironment and attenuating angiogenesis. Furthermore, the data suggest that the lack of erythroid DARC, as occurs in 70% of the African-American population, may be an important contributing factor to the increased progression of, and mortality to, prostate cancer in this population.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 25, 2005. Accepted for publication September 9, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and STAT1 in prostate cancer cells causes proangiogenic production of CXC chemokines. Am. J. Physiol. 286,C840-C847This article has been cited by other articles:
|
|
 |
|
  L. W. Horton, Y. Yu, S. Zaja-Milatovic, R. M. Strieter, and A. Richmond Opposing Roles of Murine Duffy Antigen Receptor for Chemokine and Murine CXC Chemokine Receptor-2 Receptors in Murine Melanoma Tumor Growth Cancer Res., October 15, 2007; 67(20): 9791 - 9799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Iiizumi, S. Bandyopadhyay, and K. Watabe Interaction of Duffy Antigen Receptor for Chemokines and KAI1: A Critical Step in Metastasis Suppression Cancer Res., February 15, 2007; 67(4): 1411 - 1414. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
