Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print April 12, 2007, 10.1148/radiol.2432060604
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2432060604v1
243/3/703    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kruskal, J. B.
Right arrow Articles by Goldberg, S. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kruskal, J. B.
Right arrow Articles by Goldberg, S. N.
(Radiology 2007;243:703-711.)
© RSNA, 2007

Experimental Studies

Hepatic Colorectal Cancer Metastases: Imaging Initial Steps of Formation in Mice1

Jonathan B. Kruskal, MD, PhD, Abdallah Azouz, MD, Houari Korideck, MD, Moussa El-Hallak, MD, Simon C. Robson, MD, PhD, Peter Thomas, PhD, and S. Nahum Goldberg, MD

1 From the Departments of Radiology (J.B.K., A.A., H.K., M.E., S.N.G.) and Medicine (S.C.R.), Beth Israel Deaconess Medical Center and Harvard Medical School, West Clinical Center-CC302B, 1 Deaconess Rd, Boston MA 02215; and Departments of Surgery and Biomedical Sciences, Creighton University, Omaha, Neb (P.T.). From the 2005 RSNA Annual Meeting. Received April 4, 2006; revision requested June 20; revision received July 14; final version accepted September 8. Supported by grants from the National Institutes of Health (R21 CA89634-JK, HL57307, and HL63972-SCR). J.B.K. supported by the RSNA Research and Education Foundation (RSNA Scholar Award) and the Society of Gastrointestinal Radiologists (Philip H. Meyers Research Award). Address correspondence to J.B.K. (e-mail: jkruskal{at}bidmc.harvard.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively use optical imaging to study the cell-specific mechanisms of entrapment and subsequent growth of two human colon cancer cell lines differing in their propensity to form hepatic metastases.

Materials and Methods: In this Animal Care Committee–approved study, intravital optical imaging was performed in exteriorized livers of three groups of mice after intrasplenic inoculation of human colon cancer cells. Group 1 mice (control group; n = 12) received a cell-maintaining solution only. Groups 2 and 3 (n = 12 in each) were administered poorly (MIP-101 colon cancer cells) or highly (CX-1 colon cancer cells) metastatic cells. Imaging was performed on postinoculation days 0, 1, 3, and 6 to document sites and mechanisms of tumor cell entrapment and presence and sites of endothelial cell activation and of tumor cell interactions with systemic macrophages and Kupffer cells. Fluorescence intensity of Kupffer cells was compared by using the Mann-Whitney test. Immunohistochemistry served as the reference standard for all in vivo observations.

Results: Whereas both MIP-101 and CX-1 colon cancer cells adhered to periportal Kupffer cells, the CX-1 cells resulted in Kupffer cell activation, evidenced in vivo by increased visible peroxidase activity (P < .05). Only CX-1 cells were associated with subsequent downstream endothelial cell activation, evidenced by in vivo expression of E-selectin. By day 6, regression of periportal MIP-101 tumor growth correlated with ingrowth of systemic macrophages, while CX-1 tumor growth, originating in the outflow venous regions, correlated with translobular migration and ingrowth of activated Kupffer cells.

Conclusion: Formation of hepatic colon cancer metastases is cancer cell–type specific, with cell lines differing in their mechanisms and intrahepatic locations of initial entrapment and Kupffer cell activation and subsequent growth.

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Current data suggest that resection is the treatment of choice for colorectal liver metastases, with 5-year survival rates approaching 60% (1–3). It is likely that earlier detection of hepatic metastases would make more patients eligible for treatment, and a greater understanding of mechanisms associated with the formation of liver metastases would provide other therapeutic options.

Recent advances in understanding the biology of liver metastases may facilitate imaging studies (4,5). In vitro and animal studies have revealed unique cellular and molecular events that occur during formation of liver metastases and that may potentially be imaged with molecular or cellular imaging techniques, including intravital microscopy (6–9).

The development and refinement of optical imaging techniques and contrast agents now offers the potential to directly image cellular and molecular events in the livers of live animals during the initial and earliest stages of the formation of metastases (6–9). The availability of well-characterized human colon cancer cell lines that differ in their intrinsic hepatic metastatic potential provides an ideal model for studying in vivo cellular interactions. Thus, the purpose of our study was to prospectively use optical imaging to study the cell-specific mechanisms of entrapment and subsequent growth of two human colon cancer cell lines differing in their propensity to form hepatic metastases.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Experimental Plan
Intravital microscopy was performed after intrasplenic inoculation of human colon cancer cells into two groups of mice (three mice per group studied on postinoculation days 0, 1, 3, and 6). With a control group of 12 mice, a total of 36 mice were studied. Livers from tumor-bearing live mice (12 mice per group, three mice studied per time point) were studied. All in vivo observations were correlated with results of immunohistochemical studies of the same livers, which served as the standard of reference for intravital imaging findings.

Animal Models
Two well-characterized human colon cancer cell lines that differ in their propensity for forming liver metastases were used (10). Human CX-1 colon carcinoma cells are a moderately to well-differentiated human colorectal cancer line derived from the HT-29 cell line and were provided by L. B. Chen, PhD (Dana-Farber Cancer Institute, Boston, Mass). Unlike CX-1 cells, human MIP-101 cells do not express carcinoembryonic antigen (CEA) on their cell membranes (10). All cells were maintained in RPMI 1640 medium (GIBCO Laboratories Life Technologies, Grand Island, NY) with 10% fetal bovine serum (SIGMA Chemical, St Louis, Mo), 1% L-glutamine, 100 U penicillin G per milliliter of water, and 100 mg/mL streptomycin (all from GIBCO Laboratories Life Technologies).

Experiments were performed according to a protocol approved by the institutional Animal Care and Use Committee of Harvard Medical School, in accordance with the guidelines issued by the National Institutes of Health for care of laboratory animals. All mice were housed in a pathogen-free temperature-controlled environment and were allowed access to food and water ad libitum. Inbred 6-week-old male athymic nude mice (Harlan-Sprague-Dawley, Indianapolis, Ind) that ranged in weight from 20 to 25 g were used in this study. Animal microsurgery, tumor cell injection, and optical microscopy were all performed by a single author (J.B.K.) with 15 years of such experience. Anesthesia was induced with an intraperitoneal mixture of ketamine and xylazine (Abbott Laboratories, North Chicago, Ill) at a strength of 100 mg/mL, given as a dose of 0.1 mL per 20 g of body weight. Under direct vision, cells were injected into the spleen through a mini-laparotomy incision by using a 27-gauge needle. All cells spread to the liver, where invasive and vascular metastases were established by day 7.

Three groups of mice were studied: Control mice received intrasplenic injection of 100 µL of the control solution used for maintaining all tumor cells in culture (group 1, n = 12), another group of mice were injected with 2 x 106 poorly metastasizing MIP-101 human colon cancer cells (group 2, n = 12), and another group of mice were injected with highly metastatic human CX-1 colon cancer cells (group 3, n = 12). In the latter two groups, all mice received intrasplenic inoculation of 100 µL of approximately 2 x 106 human colon cancer cells in 0.5 mL phosphate-buffered saline, pH 7.4.

Hepatic in Vivo Optical Microscopy
Dynamic in vivo microscopy was performed on exteriorized livers of all mice to 6 days after inoculation (11,12). Briefly, a standard compound trinocular microscope was modified for in vivo microscopy and equipped for transillumination and epiillumination. Images were obtained at a magnification of x100–101 500 with a modified microscope system (Optiphot; Nikon, Tokyo, Japan) and were recorded in both color (SSC-DC50A Exwave CCD camera; Sony Medical Systems, Tokyo, Japan) and black and white (CCD-72 video camera; DAGE/MTI, Michigan City, Ind). Images were captured into a computer (Dell Computers, Round Rock, Tex) by using a real-time video frame grabbing and processing graphics accelerator card (Radeon All-In-Wonder 8500 DV card; ATI Technologies, Santa Clara, Calif) for use with image analysis software (IMAGE 1.62; National Institutes of Health, Bethesda, Md [available at ftp://rsbweb.nih.gov/pub/nih-image/]). Images were also captured onto digital tapes by using a recorder (miniDV; Sony) for subsequent data and image analysis.

Image Parameters
During each experiment, a series of nine parameters were recorded in each liver (Table). These parameters were selected because (a) they are all visible and recordable in vivo, (b) they can be compared with data from similar studies performed with different cell lines, and (c) they have been shown to be related to the formation of metastases (6–9).


View this table:
[in this window]
[in a new window]

  		Parameters Recorded during Dynamic Optical Microscopy of Exteriorized Mouse Livers

 
Human colon cancer cells.—Human colon cancer cells were identified in vivo by using fluorescence microscopy after prelabeling of tumor cells with acetoxymethyl ester of calcein (Calcein-AM; Molecular Probes, Eugene, Ore) before intrasplenic inoculation (13). Calcein-AM is nonfluorescent until it is loaded into cells. Once inside cells, intracellular esterases cleave the acetoxymethyl esters to liberate the polar green–fluorescent dye calcein, which is visible with fluorescence microscopy. The dye was added to suspended or adherent cells at a final working concentration of 15 µM and was incubated for 30 minutes before washing and injection.

Macrophages.—Phagocytic intrahepatic macrophages were identified in vivo after intravenous administration of 20 µL fluorescent yellow microspheres (Fluoresbrite microparticles, 0.05 µm; Polysciences, Warrington, Pa) (6). Kupffer cell activation was documented in vivo through identification of intracellular peroxidase activity after intravenous administration of a cell-permeant fluorogenic probe (FcOxyblast; Molecular Probes). The reagent consists of anti–bovine serum albumin immunoglobulin G complexed with bovine serum albumin linked to H2DCF, a nonfluorescent fluorochrome. The immune complex binds to Fc receptors on the Kupffer cells, resulting in internalization and oxidation of H2DCF to green fluorescent dichlorofluorescein, which occurs after a nicotinamide adenine dinucleotide phosphate diaphorase oxidase–mediated oxidative burst (14).

Endothelial cells.—Sites of endothelial cell E-selectin activation and consequent expression were identified after intravenous injection of a mouse monoclonal antibody, CD62E (100 µL at 50 µg/mL) (clone 10E9.6; BD Pharmingen, San Diego, Calif), diluted in phosphate-buffered saline. A second-step reagent, streptavidin-phycoerythrin (diluted in 10 mmol/L phosphate-buffered saline, pH 7.4) (BD Pharmingen) was used to localize antibody bound to endothelial cells. Before injection of fluorochromes, all solutions were dialyzed for 24 hours in sodium-azide–free 0.15-mol/L phosphate-buffered saline (5000 mL x 3, pH 7.4, 4°C).

Immunohistochemistry
Immunohistochemistry was used to confirm our in vivo observations. At the end of each microscopic examination, livers were removed, immediately frozen with liquid nitrogen, and stored at –80°C. We prepared 0.05-mm-thick slices by using a microtome. Immunohistochemical analysis of the different antigens was performed by using an avidin-biotin complex peroxidase kit (Vectastain; Vector Laboratories, Burlingame, Calif). A standard immunohistochemical technique (avidin-biotin-peroxidase) was employed for immunolocalization of different antigens. Nonspecific binding was blocked with 10% goat serum. Liver samples were incubated with the primary monoclonal antibody, followed by incubation at room temperature with a secondary biotinylated antibody (BA-2000 and BA-9400; Vector Laboratories) and an avidin-biotin-peroxidase complex. A peroxidase DAB substrate kit, 3,3'-diaminobenzidine (SK 4100; Vector Laboratories) was used to visualize immunostaining of the CD11b, CD31, and CD62E antigens. Diaminobenzidine plus Ni2+ was used to visualize immunostaining of CEA. The substrate kit 3-amino-9-ethylcarbazole (SK 4200; Vector Laboratories) was used to visualize MOMA2, and the VECTOR VIP substrate kit (SK 4600; Vector Laboratories) was used to visualize F4/80.

Colon cancer cells were localized by using antibodies to CEA (RTU-CEA-2; Novocastra, Newcastle upon Tyne, England). Quiescent and activated systemic and resident hepatic macrophages were distinguished by using a panel of antibodies: A panspecific quiescent rat antimouse macrophage probe (for F4/80 antigen) and probes specific for activated macrophages (MOMA-2) and systemic macrophages distinct from Kupffer cells (CD11b) (15–17), respectively, were used to characterize the status and nature of macrophages adjacent to and within developing tumors. All antibodies were obtained from Accurate Chemical, Westbury, NY. Endothelial cells were identified by using rat antimouse CD31 and CD62E (both from BD Pharmingen). After exposure to 0.1% trypsin (DakoCytomation, Carpinteria, Calif) for 15 minutes and quenching with 0.3% hydrogen peroxide in methanol, slices were subjected to heat-induced antigen retrieval by boiling them for 10 minutes in citrate-buffered PH6 solution (Invitrogen, Carlsbad, Calif). They were then incubated with antibodies for 1 hour at room temperature; this was followed by standard ABC immunostaining. As positive controls for macrophages, mouse spleens were used. Replacement of the primary antibody with an equivalent concentration of nonimmune mouse immunoglobulin G (Vector Laboratories) was used for negative controls. The distribution of immunostained antigens was assessed by two observers (A.A., H.K.) who were blinded to the samples being evaluated.

Data and Statistical Analysis
Immunohistochemical images were analyzed by two authors (A.A. and H.K., with 7 and 5 years of experience, respectively, in performing and analyzing the results of immunohistochemical studies), who reached consensus about localization of cells and the presence of positive immunochemical staining. When consensus was not achieved, a third author (M.E.) with 5 years of experience served as a tie breaker. The two sources of data were the intravital videomicroscopic images and the immunohistochemical slides. Given the dynamic nature of videomicroscopy images and the use of immunochemistry to confirm dynamic observations, all data describing the presence, location, activation status, and population of cells were descriptive. Fluorescence intensity produced by activated Kupffer cells in vivo is given as mean ± standard error of the mean for all visible cells in each group. Data were initially assessed for normality with the use of normal probability plots. On the basis of these results, it was decided to use nonparametric procedures to compare data from the three groups. The Kruskal-Wallis test was used to test for overall equality of medians in each data group. When significant differences occurred, independent samples were compared by using the Mann-Whitney test at an overall significance level of .05. Analyses were performed by using software (JMP IN, version 5.1; SAS Institute, Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
All mice survived intrasplenic injection of the tumor cells and the planned delay until laparotomy and optical microscopy. Liver samples from sacrificed mice were obtained on days 0 (same day), 1, 3, and 6 (metastases are established by day 6) after tumor cell administration and were immediately frozen for subsequent immunohistochemical analysis. In the control group (group 1), no Kupffer cell or endothelial cell activation was identified.

Mechanisms and Sites of Initial Tumor Cell Entrapment
After intrasplenic inoculation, cells from both colon cancer lines reached the liver within 5–8 minutes, where two to six visible (fluorescent) static tumor cells per high-power field were identified in periportal and sinusoidal inlet regions of peripheral hepatic lobules (Fig 1). No mechanical entrapment of CX-1 tumor cells (group 2) or MIP-101 colon cancer cells (group 3) was identified. CX-1 cancer cells were identified adhering to resident Kupffer cells located in periportal zones (Fig 1). Dynamic imaging showed that adherent tumor cells from both groups caused no sinusoidal occlusion (Fig 1), and no increased leukocyte adherence was identified on sinusoidal endothelial surfaces at this early stage.


Figure 1A
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Initial entrapment of colon cancer cells in periportal zones of exteriorized mouse livers. Representative in vivo video microscopy images (original magnification, x40) obtained after intrasplenic inoculation of colon cancer cells show (a) fluorescent CX-1 colon cancer cell (arrow) and (b) MIP-101 colon cancer cell (arrow) adhering to surface of portal venules. No mechanical impaction was identified.

 

Figure 1B
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Initial entrapment of colon cancer cells in periportal zones of exteriorized mouse livers. Representative in vivo video microscopy images (original magnification, x40) obtained after intrasplenic inoculation of colon cancer cells show (a) fluorescent CX-1 colon cancer cell (arrow) and (b) MIP-101 colon cancer cell (arrow) adhering to surface of portal venules. No mechanical impaction was identified.

 
Interactions between Tumor Cells and Macrophages
Unlike mice in the control group, in all mice inoculated with CX-1 cells (group 2), imaging 24 hours after tumor cell inoculation showed activation of Kupffer cells in periportal zones of the liver (Fig 2). In mice that received poorly metastatic MIP-101 colon cancer cells, Kupffer cells became activated, but the extent of Kupffer cell peroxidase activity (median fluorescence intensity, 71.9 arbitrary fluorescence units ± 15.2) was less (P < .05) than in mice inoculated with CX-1 cells (median fluorescence intensity, 126.8 arbitrary fluorescence units ± 16.2) (Fig 2).


Figure 2A
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Activation of periportal hepatic Kupffer cells by colon cancer cells. (a, b) Video microscopy images obtained in exteriorized livers of mice inoculated with (a) CX-1 or (b) MIP-101 colon cancer cells 1 day previously show fluorescent activity in periportal Kupffer cells. Note lower level of fluorescence in b than in a. These fluorescence images show intracellular peroxidase activity after injection of a probe engineered to fluoresce when activated by peroxidase. (c) Location and presence of Kupffer cell activation was confirmed with immunohistochemistry performed by using antibodies to the panspecific macrophage activation marker MOMA-2. Note tumor cells (arrow) adjacent to wall of this portal venule (PV). (MOMA-2 stain; original magnification, x20.)

 

Figure 2B
View larger version (46K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Activation of periportal hepatic Kupffer cells by colon cancer cells. (a, b) Video microscopy images obtained in exteriorized livers of mice inoculated with (a) CX-1 or (b) MIP-101 colon cancer cells 1 day previously show fluorescent activity in periportal Kupffer cells. Note lower level of fluorescence in b than in a. These fluorescence images show intracellular peroxidase activity after injection of a probe engineered to fluoresce when activated by peroxidase. (c) Location and presence of Kupffer cell activation was confirmed with immunohistochemistry performed by using antibodies to the panspecific macrophage activation marker MOMA-2. Note tumor cells (arrow) adjacent to wall of this portal venule (PV). (MOMA-2 stain; original magnification, x20.)

 

Figure 2C
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Activation of periportal hepatic Kupffer cells by colon cancer cells. (a, b) Video microscopy images obtained in exteriorized livers of mice inoculated with (a) CX-1 or (b) MIP-101 colon cancer cells 1 day previously show fluorescent activity in periportal Kupffer cells. Note lower level of fluorescence in b than in a. These fluorescence images show intracellular peroxidase activity after injection of a probe engineered to fluoresce when activated by peroxidase. (c) Location and presence of Kupffer cell activation was confirmed with immunohistochemistry performed by using antibodies to the panspecific macrophage activation marker MOMA-2. Note tumor cells (arrow) adjacent to wall of this portal venule (PV). (MOMA-2 stain; original magnification, x20.)

 
Endothelial Cell Activation
After intrasplenic injection of MIP-101 colon cancer cells and in control mouse livers, no fluorescent antibody bound to E-selectin on sinusoidal or outflow endothelial cells in vivo (Fig 3). In mice that received CX-1 cancer cells, visible expression of E-selectin occurred within 20 minutes of tumor cell injection and persisted for the 6-day duration of data collection (Fig 3). This new expression of E-selectin occurred on the surface of endothelial cells located in sinusoidal outlet and postsinusoidal regions. All data were confirmed on immunohistochemical slices prepared from the same livers (Fig 3). The spatial distribution of expression of E-selectin occurred in the same locations as enhanced leukocyte endothelial adherence to the lining of outflow sinusoids and postsinusoidal venules (Fig 4).


Figure 3A
View larger version (58K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Indirect imaging of E-selectin expression on outflow sinusoidal endothelial cells. (a, b) Video microscopic images obtained after intravenous injection of fluorescent-labeled antibodies to E-selectin. (a) Image shows representative example of binding of anti–E-selectin antibody to sinusoidal outflow endothelial cells in mice administered CX-1 colon cancer cells. (b) Note absence of fluorescence in mice inoculated with MIP-101 cancer cells. (c) Results of immunohistochemical studies confirmed presence of E-selectin on endothelial cells lining outflow sinusoids in mice administered CX-1 tumor cells. CV = central venule. (E-selectin on hematoxylin-eosin stain; original magnification, x20.)

 

Figure 3B
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Indirect imaging of E-selectin expression on outflow sinusoidal endothelial cells. (a, b) Video microscopic images obtained after intravenous injection of fluorescent-labeled antibodies to E-selectin. (a) Image shows representative example of binding of anti–E-selectin antibody to sinusoidal outflow endothelial cells in mice administered CX-1 colon cancer cells. (b) Note absence of fluorescence in mice inoculated with MIP-101 cancer cells. (c) Results of immunohistochemical studies confirmed presence of E-selectin on endothelial cells lining outflow sinusoids in mice administered CX-1 tumor cells. CV = central venule. (E-selectin on hematoxylin-eosin stain; original magnification, x20.)

 

Figure 3C
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: Indirect imaging of E-selectin expression on outflow sinusoidal endothelial cells. (a, b) Video microscopic images obtained after intravenous injection of fluorescent-labeled antibodies to E-selectin. (a) Image shows representative example of binding of anti–E-selectin antibody to sinusoidal outflow endothelial cells in mice administered CX-1 colon cancer cells. (b) Note absence of fluorescence in mice inoculated with MIP-101 cancer cells. (c) Results of immunohistochemical studies confirmed presence of E-selectin on endothelial cells lining outflow sinusoids in mice administered CX-1 tumor cells. CV = central venule. (E-selectin on hematoxylin-eosin stain; original magnification, x20.)

 

Figure 4A
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Colocalization of sites of E-selectin expression and concurrent leukocyte adherence. (a, b) Video microscopic images obtained in exteriorized liver of mouse inoculated with CX-1 colon cancer cells 3 days previously show adherence of fluorescent leukocytes to outflow endothelial cells (arrows). In vivo binding of fluorescent antibody to E-selectin is demonstrated on same outflow endothelial cells in b.

 

Figure 4B
View larger version (91K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Colocalization of sites of E-selectin expression and concurrent leukocyte adherence. (a, b) Video microscopic images obtained in exteriorized liver of mouse inoculated with CX-1 colon cancer cells 3 days previously show adherence of fluorescent leukocytes to outflow endothelial cells (arrows). In vivo binding of fluorescent antibody to E-selectin is demonstrated on same outflow endothelial cells in b.

 
Sites and Mechanism of Earliest Tumor Cell Growth
Intravital imaging of mouse livers 1–3 days after CX-1 tumor cell inoculation demonstrated arrest and transendothelial migration of tumor cells in the sinusoidal outflow but not periportal inflow regions (Fig 5). CX-1 cancer cell arrest occurred in the same sites as those where E-selectin expression was identified. Activated Kupffer cells surrounded the CX-1 cells that originally adhered to periportal vessels (Fig 6). Whereas no MIP-101 cells (group 3) were identified in these outflow regions, MIP-101 cells located in periportal inflow zones became surrounded by activated systemic macrophages (expressing MOMA-2 and CD11b) rather than and distinct from CD11b-negative resident hepatic Kupffer cells (Fig 6).


Figure 5A
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Arrest of CX-1 colon cancer cells in pericentral zone of mouse hepatic lobule. (a–c) Representative video microscopic images obtained from exteriorized livers of mice inoculated with CX-1 colon cancer cells show (a) early adherence (arrows) to (b) luminal surface of central vein followed by growth (arrow) and (c) progressive surrounding of outflow venule (arrows). (d) Results of immunohistochemistry showed growth of CEA-positive CX-1 tumor cells (small arrow) adjacent to central venule (CV). (CEA stain; original magnification, x20.)

 

Figure 5B
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Arrest of CX-1 colon cancer cells in pericentral zone of mouse hepatic lobule. (a–c) Representative video microscopic images obtained from exteriorized livers of mice inoculated with CX-1 colon cancer cells show (a) early adherence (arrows) to (b) luminal surface of central vein followed by growth (arrow) and (c) progressive surrounding of outflow venule (arrows). (d) Results of immunohistochemistry showed growth of CEA-positive CX-1 tumor cells (small arrow) adjacent to central venule (CV). (CEA stain; original magnification, x20.)

 

Figure 5C
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c: Arrest of CX-1 colon cancer cells in pericentral zone of mouse hepatic lobule. (a–c) Representative video microscopic images obtained from exteriorized livers of mice inoculated with CX-1 colon cancer cells show (a) early adherence (arrows) to (b) luminal surface of central vein followed by growth (arrow) and (c) progressive surrounding of outflow venule (arrows). (d) Results of immunohistochemistry showed growth of CEA-positive CX-1 tumor cells (small arrow) adjacent to central venule (CV). (CEA stain; original magnification, x20.)

 

Figure 5D
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d: Arrest of CX-1 colon cancer cells in pericentral zone of mouse hepatic lobule. (a–c) Representative video microscopic images obtained from exteriorized livers of mice inoculated with CX-1 colon cancer cells show (a) early adherence (arrows) to (b) luminal surface of central vein followed by growth (arrow) and (c) progressive surrounding of outflow venule (arrows). (d) Results of immunohistochemistry showed growth of CEA-positive CX-1 tumor cells (small arrow) adjacent to central venule (CV). (CEA stain; original magnification, x20.)

 

Figure 6A
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a: Imaging of interactions between colon cancer cells and macrophages. (a) In vivo video microscopic image obtained 5 days after administration of CX-1 colon cancer cells shows presence of phagocytic macrophages surrounding growing tumor (T). (b, c) In livers of mice administered CX-1 colon cancer cells, results of immunohistochemistry show activated MOMA-2–positive Kupffer cells in (b) periportal zone (PV) on day 2 (original magnification, x20) and (c) outflow regions of pericentral zones adjacent to a central venule (CV) by day 5 (original magnification, x20). (d) In mice inoculated with MIP-101 cells, immunohistochemistry confirms presence of MOMA-2–positive activated systemic macrophages (Cd11b positive) in and adjacent to tumors, which are located adjacent to a portal venule (PV). (Original magnification, x60.)

 

Figure 6B
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b: Imaging of interactions between colon cancer cells and macrophages. (a) In vivo video microscopic image obtained 5 days after administration of CX-1 colon cancer cells shows presence of phagocytic macrophages surrounding growing tumor (T). (b, c) In livers of mice administered CX-1 colon cancer cells, results of immunohistochemistry show activated MOMA-2–positive Kupffer cells in (b) periportal zone (PV) on day 2 (original magnification, x20) and (c) outflow regions of pericentral zones adjacent to a central venule (CV) by day 5 (original magnification, x20). (d) In mice inoculated with MIP-101 cells, immunohistochemistry confirms presence of MOMA-2–positive activated systemic macrophages (Cd11b positive) in and adjacent to tumors, which are located adjacent to a portal venule (PV). (Original magnification, x60.)

 

Figure 6C
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6c: Imaging of interactions between colon cancer cells and macrophages. (a) In vivo video microscopic image obtained 5 days after administration of CX-1 colon cancer cells shows presence of phagocytic macrophages surrounding growing tumor (T). (b, c) In livers of mice administered CX-1 colon cancer cells, results of immunohistochemistry show activated MOMA-2–positive Kupffer cells in (b) periportal zone (PV) on day 2 (original magnification, x20) and (c) outflow regions of pericentral zones adjacent to a central venule (CV) by day 5 (original magnification, x20). (d) In mice inoculated with MIP-101 cells, immunohistochemistry confirms presence of MOMA-2–positive activated systemic macrophages (Cd11b positive) in and adjacent to tumors, which are located adjacent to a portal venule (PV). (Original magnification, x60.)

 

Figure 6D
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6d: Imaging of interactions between colon cancer cells and macrophages. (a) In vivo video microscopic image obtained 5 days after administration of CX-1 colon cancer cells shows presence of phagocytic macrophages surrounding growing tumor (T). (b, c) In livers of mice administered CX-1 colon cancer cells, results of immunohistochemistry show activated MOMA-2–positive Kupffer cells in (b) periportal zone (PV) on day 2 (original magnification, x20) and (c) outflow regions of pericentral zones adjacent to a central venule (CV) by day 5 (original magnification, x20). (d) In mice inoculated with MIP-101 cells, immunohistochemistry confirms presence of MOMA-2–positive activated systemic macrophages (Cd11b positive) in and adjacent to tumors, which are located adjacent to a portal venule (PV). (Original magnification, x60.)

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
Optical imaging has allowed us to verify results from previous in vitro studies (18) and, more importantly, to demonstrate different mechanisms and sites of early intrahepatic cancer cell growth for two different colon cancer cell lines. Unlike currently available clinical imaging modalities, optical and related imaging techniques permit evaluation of physiologic and molecular events in animal models.

Mechanisms and Sites of Initial Tumor Cell Entrapment
Unlike the expected mechanical entrapment of circulating tumor cells by Kupffer cells in the liver (18–20), mechanical entrapment of CX-1 tumor cells was not observed in our mice. Results of our in vivo study demonstrate that activation of Kupffer cells also occurs and is tumor cell type specific. Only the more aggressive metastatic CX-1 colon cell line was associated with Kupffer cell activation in vivo. Our data are further supported by the results of Khatib et al (21), who showed that CX-1 cells trigger a series of proinflammatory responses in the liver. More specifically, the influx of CX-1 cells results in tumor necrosis factor-{alpha} production by activated Kupffer cells, followed by expression of E-selectin, similar to what we have shown with intravital imaging. Whether free or membrane-bound, CEA induces brisk activation of Kupffer cells that results in the release of cytokines (22) through a receptor system that is related to the RNA binding protein hnRNP M4 (23,24). CX-1 cells secrete and express CEA antigen on their surface, while MIP-101 cells do not secrete or express CEA (25). Results of previous studies (26) have shown that secreted CEA is the most important factor in the activation of Kupffer cells. The presence of cell surface CEA on the tumor cells does not seem to play a role in tumor cell binding to the endothelium (26). The secretion of CEA by CX-1 cells thus provides a logical explanation for the in vivo activation of periportal Kupffer cells that we observed.

Endothelial Cell Activation
In addition to arresting populations of circulating tumor cells, Kupffer cells also control early growth of hepatic metastases by clearing cancer cells from the liver sinusoids (27). Kupffer cells also modulate the immune response to cancer cells in the liver (28) and, through expression and release of cytokines, activate endothelial cells in vitro, inducing expression of E-selectin (22). Our study showed in vivo expression of E-selectin by sinusoidal outlet endothelial cells that occurred following upstream Kupffer cell activation and preceded visible binding of tumor cells to the outflow endothelial surfaces. This sequential stepwise activation of different cell populations in the liver may be essential for the subsequent entrapment of circulating tumor cells in the liver. It is known that activation of downstream sinusoidal and postsinusoidal endothelial cells, with expression of E-selectin, precedes adherence of circulating tumor cells to the endothelial cells (29–31). The role of E-selectin in enhancing tumor cell tethering and arrest has been studied in vitro (31). Colon cancer cells adhere to activated E-selectin–expressing endothelial cells in vitro (29) and require expression of endothelial E-selectin for transendothelial migration (28,32). Binding of colon cancer cells to E-selectin is mediated by the sialyl Lewis X glycoprotein present on certain cancer cell membranes (33).

In the clinical setting, serum levels of soluble E-selectin are strongly associated with the clinical course of metastatic disease (34) and endothelial cell activation. Interestingly, colon cancer metastases have been inhibited in an animal model by administration of soluble E-selectin that binds to and blocks the cell membrane glycoprotein (35). Khatib et al (36) have used antisense oliginucleotides to C-raf to block endothelial cell expression of E-selectin and have shown decreased adherence of CX-1 cells in vitro and a reduction in liver metastases from these cells in vivo. Of relevance to our study, CX-1 colon cancer cells express sialyl Lewis X but MIP-101 cells do not; this may well explain their different abilities to adhere to endothelial surfaces and different metastatic properties (32,37).

Sites of Earliest Tumor Cell Growth
We have shown that expression of E-selectin only occurred after CX-1 tumor cell inoculation and that, unlike MIP-101 colon cancer cells, CX-1 cancer cells adhered to endothelial cells expressing E-selectin. Adhesion of colorectal cancer cells to endothelium is mediated by cytokines from CEA-stimulated Kupffer cells (23). The mechanisms of early formation of hepatic colorectal cancer metastases thus appear to be cancer cell type specific, an observation that may have therapeutic implications. Optical imaging has allowed us to confirm in vitro data showing that the mechanisms of tumor cell adhesion depend in part on specific interactions occurring between membrane receptors expressed on both tumor cells and activated endothelial cells (38).

Macrophage Interactions
We have demonstrated that systemic macrophages rather than Kupffer cells are recruited to periportal sites of early adherence and growth of the poorly metastatic MIP-101 cells. Moreover, the increased presence of systemic macrophages correlates strongly with tumor cell regression. These observations are in contradistinction to activated Kupffer cells that migrate to pericentral zones toward sites of CX-1 tumor cell growth. These data suggest that Kupffer cells support the growth of hepatic metastases, whereas systemic macrophages inhibit growth of the MIP-101 cell line. Results of other histology-based studies (39) have shown ingrowth of Kupffer cells into developing colorectal cancer metastases. It is possible that other tumor cells may interact differently with different macrophage populations.

Once tumor cells have commenced growth in the liver, systemic and resident macrophages appear to play an important role in inhibiting and supporting tumor growth, respectively. Characterization of the origin and activation status of these macrophage populations is important because some macrophages stimulate tumor growth and angiogenesis (28), whereas others may play an inhibitory role (40,41). Previous histology-based studies (40,42) have documented the roles of macrophages in surveillance of hepatic colorectal cancer metastases. Given the potential therapeutic importance of these observations, further validation and comparison with other tumor cell lines is warranted.

Results of in vitro studies (42) have shown that different macrophage populations, including systemic macrophages and Kupffer cells, are involved in the immune response for or against developing hepatic metastases. Unlike in vitro histologic studies, intravital microscopy permits identification and localization of intrahepatic macrophages and can be used to quantitate macrophage phagocytic function and activation status. Our study results have shown that, unlike MIP-101 cells that become surrounded by systemic macrophages, growing CX-1 metastases are surrounded and most likely supported by Kupffer cells recruited from the periportal zone.

Study Limitations
There were possible limitations of our study. First, one cannot necessarily translate animal data to the human clinical setting. Whereas the nude mouse model is very practical for studying human tumor xenografts, these rodents are T-cell deficient, and, thus, any contribution of T cells to formation of liver metastases cannot be studied with this model. Second, the cell-specific variations in the human cancer cell lines used in our study should be considered. It is possible that human cells interact differently from mouse tumor cells during formation of metastases, providing further justification for human studies. These data should thus be explored in human studies.

Practical applications: If our data can be confirmed, we will have provided evidence of several potential new diagnostic and therapeutic targets, as well as opportunities for imaging to predict prognosis and to select therapeutic options. Promising biologic agents, including peptide mimetics, carbohydrate antagonists, and antisense oligonucleotides, have been developed that enhance the tumoricidal effects of Kupffer cells (43–45), that block expression of E-selectin (36,46), or that inhibit tumor cell binding to E-selectin through blockage of sialyl Lewis X (47,48). Developments in the use of small interfering RNA to inhibit gene function (49) will also be powerful tools in the creation of novel therapies to prevent and treat hepatic metastasis from colorectal cancers.

In summary, using an intravital optical imaging system, we have documented different mechanisms responsible for the initial entrapment and subsequent growth of two different human colon cancer cell lines metastasizing to the mouse liver. Elucidating the causes for these differences could have important clinical implications.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: CEA = carcinoembryonic antigen

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, J.B.K.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, J.B.K., A.A., M.E., P.T.; experimental studies, J.B.K., A.A. H.K., M.E., S.C.R.; statistical analysis, J.B.K., S.C.R., S.N.G.; and manuscript editing, J.B.K., A.A., S.C.R., P.T., S.N.G.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004;239:818–825.[CrossRef][Medline]
  2. Gillams AR, Lees WR. Radio-frequency ablation of colorectal liver metastases in 167 patients. Eur Radiol 2004;14:2261–2267.[CrossRef][Medline]
  3. Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002;235:759–765.[CrossRef][Medline]
  4. Ahmad SA, Berman RS, Ellis LM. Biology of colorectal liver metastases. Surg Oncol Clin N Am 2003;12:135–150.[CrossRef][Medline]
  5. Bogenrieder T, Herlyn M. Axis of evil: molecular mechanisms of cancer metastasis. Oncogene 2003;22:6524–6536.[CrossRef][Medline]
  6. Ishii S, Mizoi T, Kawano K, et al. Implantation of human colorectal carcinoma cells in the liver studied by in vivo fluorescence videomicroscopy. Clin Exp Metastasis 1996;14:153–164.[CrossRef][Medline]
  7. Groom AC, MacDonald IC, Schmidt EE, Morris VL, Chambers AF. Tumour metastasis to the liver, and the roles of proteinases and adhesion molecules: new concepts from in vivo videomicroscopy. Can J Gastroenterol 1999;13:733–743.[Medline]
  8. Morris VL, Schmidt EE, MacDonald IC, Groom AC, Chambers AF. Sequential steps in hematogenous metastasis of cancer cells studied by in vivo videomicroscopy. Invasion Metastasis 1997;17:281–296.[Medline]
  9. Naumov GN, Wilson SM, MacDonald IC, et al. Cellular expression of green fluorescent protein, coupled with high-resolution in vivo videomicroscopy, to monitor steps in tumor metastasis. J Cell Sci 1999;112:1835–1842.[Abstract]
  10. Zimmer R, Thomas P. Expression profiling and interferon-beta regulation of liver metastases in colorectal cancer cells. Clin Exp Metastasis 2002;19:541–550.[CrossRef][Medline]
  11. Kruskal JB, Thomas P, Kane RA, Goldberg SN. Hepatic perfusion changes in mice livers with developing colorectal cancer metastases. Radiology 2004;231:482–490.[Abstract/Free Full Text]
  12. Kruskal JB, Thomas P, Nasser I, Cay O, Kane RA. Hepatic colon cancer metastases in mice: dynamic in vivo correlation with hypoechoic rims visible at US. Radiology 2000;215:852–857.[Abstract/Free Full Text]
  13. Guckelberger O, Sun XF, Sevigny J, et al. Beneficial effects of CD39/ecto-nucleoside triphosphate diphosphohydrolase-1 in murine intestinal ischemia-reperfusion injury. Thromb Haemost 2004;91:576–586.[Medline]
  14. Ryan TC, Weil GJ, Newburger PE, Haugland R, Simons ER. Measurement of superoxide release in the phagovacuoles of immune complex-stimulated human neutrophils. J Immunol Methods 1990;130:223–233.[CrossRef][Medline]
  15. Maier S, Traeger T, Entleutner M, et al. Cecal ligation and puncture versus colon ascendens stent peritonitis: two distinct animal models for polymicrobial sepsis. Shock 2004;21:505–512.[CrossRef][Medline]
  16. Schaller E, Macfarlane AJ, Rupec RA, Gordon S, McKnight AJ, Pfeffer K. Inactivation of the F4/80 glycoprotein in the mouse germ line. Mol Cell Biol 2002;22:8035–8043.[Abstract/Free Full Text]
  17. Kraal G, Rep M, Janse M. Macrophages in T and B cell compartments and other tissue macrophages recognized by monoclonal antibody MOMA-2: an immunohistochemical study. Scand J Immunol 1987;26:653–661.[CrossRef][Medline]
  18. Barbera-Guillem E, Alonso-Varona A, Vidal-Vanaclocha F. Selective implantation and growth in rats and mice of experimental liver metastasis in acinar zone one. Cancer Res 1989;49:4003–4010.[Abstract/Free Full Text]
  19. Morris VL, MacDonald IC, Koop S, Schmidt EE, Chambers AF, Groom AC. Early interactions of cancer cells with the microvasculature in mouse liver and muscle during hematogenous metastasis: videomicroscopic analysis. Clin Exp Metastasis 1993;11:377–390.[CrossRef][Medline]
  20. Mook OR, Van Marle J, Vreeling-Sindelarova H, Jonges R, Frederiks WM, Van Noorden CJ. Visualization of early events in tumor formation of eGFP-transfected rat colon cancer cells in liver. Hepatology 2003;38:295–304.[Medline]
  21. Khatib AM, Auguste P, Fallavollita L, et al. Characterization of the host proinflammatory response to tumor cells during the initial stages of liver metastasis. Am J Pathol 2005;167:749–759.[Abstract/Free Full Text]
  22. Gangopadhyay A, Bajenova O, Kelly TM, Thomas P. Carcinoembryonic antigen induces cytokine expression in Kupffer cells: implications for hepatic metastasis from colorectal cancer. Cancer Res 1996;56:4805–4810.[Abstract/Free Full Text]
  23. Bajenova OV, Zimmer R, Stolper E, Salisbury-Rowswell J, Nanji A, Thomas P. HnRNP M4 is a receptor for carcinoembryonic antigen in Kupffer cells. J Biol Chem 2001;276:31067–31073.[Abstract/Free Full Text]
  24. Bajenova O, Stolper E, Gapon S, Sundina N, Zimmer R, Thomas P. Surface expression of HnRNP M4 nuclear protein on Kupffer cells relates to its function as a carcinoembryonic antigen receptor. Exp Cell Res 2003;291:228–241.[CrossRef][Medline]
  25. Gangopadhyay A, Lazure DA, Thomas P. Adhesion of colorectal carcinoma cells to the endothelium is mediated by cytokines from CEA stimulated Kupffer cells. Clin Exp Metastasis 1998;16:703–712.[CrossRef][Medline]
  26. Jessup JM, Ishii S, Mitzoi T, Edmiston KH, Shoji Y. Carcinoembryonic antigen facilitates experimental metastasis through a mechanism that does not involve adhesion to liver cells. Clin Exp Metastasis 1999;17:481–488.[CrossRef][Medline]
  27. Timmers M, Vekemans K, Vermijlen D, et al. Interactions between rat colon carcinoma cells and Kupffer cells during the onset of hepatic metastasis. Int J Cancer 2004;112:793–802.[CrossRef][Medline]
  28. Bayon LG, Izquierdo MA, Sirovich I, et al. Role of Kupffer cells in arresting circulating tumor cells and controlling metastatic growth in the liver. Hepatology 1996;23:1224–1231.[CrossRef][Medline]
  29. Laferriere J, Houle F, Huot J. Adhesion of HT-29 colon carcinoma cells to endothelial cells requires sequential events involving E-selectin and integrin beta4. Clin Exp Metastasis 2004;21:257–264.[CrossRef][Medline]
  30. Laferriere J, Houle F, Taher MM, et al. Transendothelial migration of colon carcinoma cells requires expression of E-selectin by endothelial cells and activation of stress-activated protein kinase-2 (SAPK2/p38) in the tumor cells. J Biol Chem 2001;276:33762–33772.[Abstract/Free Full Text]
  31. Kitayama J, Tsuno N, Sunami E, et al. E-selectin can mediate the arrest type of adhesion of colon cancer cells under physiological shear flow. Eur J Cancer 2000;36:121–127.[CrossRef][Medline]
  32. Brodt P, Fallavollita L, Bresalier RS, et al. Liver endothelial E-selectin mediates carcinoma cell adhesion and promotes liver metastasis. Int J Cancer 1997;71:612–619.[CrossRef][Medline]
  33. Tomlinson J, Wang JL, Barsky SH, et al. Human colon cancer cells express multiple glycoprotein ligands for E-selectin. Int J Oncol 2000;16:347–353.[Medline]
  34. Eichbaum MH, de Rossi TM, Kaul S, Bastert G. Serum levels of soluble E-selectin are associated with the clinical course of metastatic disease in patients with liver metastases from breast cancer. Oncol Res 2004;14:603–610.[Medline]
  35. Mannori G, Santoro D, Carter L, et al. Inhibition of colon carcinoma cell lung colony formation by a soluble form of E-selectin. Am J Pathol 1997;151:233–243.[Abstract]
  36. Khatib AM, Fallavollita L, Wancewicz EV, Monia BP, Brodt P. Inhibition of hepatic endothelial E-selectin expression by C-raf antisense oligonucleotides blocks colorectal carcinoma liver metastasis. Cancer Res 2002;62:5393–5398.[Abstract/Free Full Text]
  37. Opolski A, Laskowska A, Madej J, et al. Metastatic potential of human CX-1 colon adenocarcinoma cells is dependent on the expression of sialosyl Le(a) antigen. Clin Exp Metastasis 1998;16:673–681.[CrossRef][Medline]
  38. Ding L, Sunamura M, Kodama T, et al. In vivo evaluation of the early events associated with liver metastasis of circulating cancer cells. Br J Cancer 2001;85:431–438.[CrossRef][Medline]
  39. Higashi N, Ishii H, Fujiwara T, et al. Redistribution of fibroblasts and macrophages as micrometastases develop into established liver metastases. Clin Exp Metastasis 2002;19:631–638.[CrossRef][Medline]
  40. Whitworth PW, Pak CC, Esgro J, Kleinerman ES, Fidler IJ. Macrophages and cancer. Cancer Metastasis Rev 1990;8:319–351.[CrossRef][Medline]
  41. Liu K, He X, Lei XZ, et al. Pathomorphological study on location and distribution of Kupffer cells in hepatocellular carcinoma. World J Gastroenterol 2003;9:1946–1949.[Medline]
  42. Heuff G, van der Ende MB, Boutkan H, et al. Macrophage populations in different stages of induced hepatic metastases in rats: an immunohistochemical analysis. Scand J Immunol 1993;38:10–16.[CrossRef][Medline]
  43. Liu Q, Klintman D, Corbascio M, et al. Linomide and antibody-targeted superantigen therapy abolishes formation of liver metastases in mice. Eur Surg Res 2003;35:457–463.[CrossRef][Medline]
  44. Karpoff HM, Jarnagin W, Delman K, Fong Y. Regional muramyl tripeptide phosphatidylethanolamine administration enhances hepatic immune function and tumor surveillance. Surgery 2000;128:213–218.[CrossRef][Medline]
  45. van de Water B, van Berkel TJ, Kuiper J. Activation of rat Kupffer cells to tumoricidal cells by the immunomodulator muramyl tripeptide-phosphatidylethanolamine incorporated into the novel drug carrier lactosylated low density lipoprotein. Mol Pharmacol 1994;45:971–977.[Abstract]
  46. Nubel T, Dippold W, Kleinert H, et al. Lovastatin inhibits Rho-regulated expression of E-selectin by TNFalpha and attenuates tumor cell adhesion. FASEB J 2004;18:140–142.[Abstract/Free Full Text]
  47. Fukami A, Iijima K, Hayashi M, Komiyama K, Omura S. Macrosphelide B suppressed metastasis through inhibition of adhesion of sLe(x)/E-selectin molecules. Biochem Biophys Res Commun 2002;291:1065-1070.[CrossRef][Medline]
  48. Fukuda MN, Ohyama C, Lowitz K, et al. A peptide mimic of E-selectin ligand inhibits sialyl Lewis X-dependent lung colonization of tumor cells. Cancer Res 2000;60:450–456.[Abstract/Free Full Text]
  49. Hannon GJ. RNA interference. Nature 2002;418:244–251.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2432060604v1
243/3/703    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kruskal, J. B.
Right arrow Articles by Goldberg, S. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kruskal, J. B.
Right arrow Articles by Goldberg, S. N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE