We have previously shown that a stromal cell–derived factor-1 (SDF-1; CXCL12)/CXCR4 system is involved in the establishment of lymph node metastasis, but not in that of distant metastasis, in oral squamous cell carcinoma (SCC). In this study, we investigated the role of the autocrine SDF-1/CXCR4 system, with a focus on distant metastasis in oral SCC cells. The immunohistochemical staining of SDF-1 and CXCR4 using primary oral SCCs and metastatic lymph nodes showed a significantly higher number of SDF-1–positive cases among the metastatic lymph nodes than among the primary oral SCCs, which was associated with a poor survival rate among those of the former group. The forced expression of SDF-1 in B88 cells, which exhibit functional CXCR4 and lymph node metastatic potential (i.e., the autocrine SDF-1/CXCR4 system), conferred enhanced cell motility and anchorage-independent growth potential onto the cells. Orthotopic inoculation of the transfectant into nude mice was associated with an increase in the number of metastatic lymph nodes and more aggressive metastatic foci in the lymph nodes. Furthermore, the SDF-1 transfectant (i.e., the autocrine SDF-1/CXCR4 system) exhibited dramatic metastasis to the lung after i.v. inoculation, whereas the mock transfectant (i.e., the paracrine SDF-1/CXCR4 system) did not. Under the present conditions, AMD3100, a CXCR4 antagonist, significantly inhibited the lung metastasis of the SDF-1 transfectant, ameliorated body weight loss, and improved the survival rate of tumor-bearing nude mice. These results suggested that, in cases of oral SCC, the paracrine SDF-1/CXCR4 system potentiates lymph node metastasis, but distant metastasis might require the autocrine SDF-1/CXCR4 system. (Mol Cancer Res 2007;5(7):685–94)

Chemokines are a large family of small (7-15 kDa), structurally related heparin-binding proteins that have been identified as attractants of different types of blood leukocytes to sites of infection and inflammation (1, 2). They are produced locally in the tissues and act on leukocytes through selective membrane-bound G-protein–coupled receptors, the two major subfamilies of which have been designated as CCR and CXCR. Among these chemokines and their receptors, the stromal cell–derived factor-1 (SDF-1; also referred to as the CXCL12)/CXCR4 system has been shown to be involved primarily in the site-specific distant metastasis of several types of cancer (3-9). We have reported that SDF-1 produced in lymph nodes exerts an attractive force on lymph node metastasis in CXCR4-expressing oral squamous cell carcinoma (SCC; refs. 10-12), and we also showed frequent expression of CXCR4 in ∼60% of oral SCCs at the primary site (13). Moreover, as regards cases of head and neck SCCs, including those in the oral, nasopharyngeal, and esophageal regions, several investigators have shown that this system is involved in lymph node metastasis (14-17). However, a recent report noted that a SDF-1/CXCR4 gradient regulated primarily hematogenous distant metastasis rather than lymph node metastasis (18). Clinically, oral SCCs are known to frequently metastasize to the lymph nodes, but rarely to the distant organs; the rate of the latter type of metastasis has been reported to be <20% of oral SCC (19-21). Based on these observations, it is likely that a SDF-1/CXCR4 gradient might regulate lymph node metastasis, but only in certain types of cancer such as head and neck SCCs; however, the reasons for which head and neck SCCs do not metastasize to the distant organs, despite the frequent expression of CXCR4, remains largely unknown.

One possible explanation for this phenomenon is that ectopic growth potential at distant sites in oral SCC is weaker than that in other types of cancer, even in cases when the cancer cells are able to reach the distant organs via the SDF-1/CXCR4 gradient. Previously, it has been suggested that at early stages of metastatic progression, paracrine growth mechanisms might dominate growth signals, whereas at later stages of metastatic progression, autocrine growth mechanisms might dominate the growth responses of metastatic cells (22, 23). Recently, several growth factors and their receptor systems have been suggested to contribute via an autocrine mechanism to distant metastatic spread in several types of cancer (24-28). As described above, the SDF-1 produced by metastatic sites plays an attractive and paracrine role in the cancer cells with CXCR4 expression; however, the autocrine role played by this system in distant metastasis remains unknown. In this study, we examined whether the autocrine SDF-1/CXCR4 system contributes to the distant metastasis of oral SCC.

Expression of SDF-1 and CXCR4 in the Metastatic Lymph Nodes of Patients with Oral SCC

In primary oral SCC tissue samples, SDF-1 expression was frequently detected in normal and cancer stromal cells, but was infrequently detected in cancer cells (13). For the establishment of distant metastasis, cancer cells must intravasate into (newly synthesized) micro–blood vessels or flow out from the metastatic lymph nodes. As regards cases of oral SCCs, the latter mechanism is thought to be critical because patients with distant metastases had high incidence (at the rate of 93%) of lymph node metastases (21). Thus, considering the hypothesis of the contribution of autocrine SDF-1/CXCR4 signaling on the distant metastatic spread of oral SCC, we first examined the expression of SDF-1 and CXCR4 proteins in primary oral SCC cells and also in metastatic cancer cells of lymph nodes (Fig. 1Aa-Ad). Thirty cases of oral SCC with metastatic lymph nodes were examined, and the expression of CXCR4 was found to be almost equivalent in the primary tumors and in the metastatic lymph nodes (Table 1). However, as regards SDF-1, the rate of positive expression in the metastatic lymph nodes (16 of 30) was significantly higher than that in the primary tissues (2 of 30; Table 1; P < 0.001, Wilcoxon signed rank test). Among these samples, some spindle cancer cells expressing both CXCR4 and SDF-1 were detected (Fig. 1Ac and Ad). The number of SDF-1– and CXCR4-positive metastatic lymph nodes was 14 out of a total of 30 samples. To confirm the specificity of these immunohistochemical results, we did immunohistochemistry, using metastatic lymph nodes formed in nude mice inoculated with mock or SDF-1 transfectants described in Fig. 2. SDF-1 staining could be dramatically detected only in the SDF-1 transfectant (data not shown). Thus, these results indicated that approximately half of the cases of oral SCC with metastatic lymph nodes had acquired autocrine SDF-1/CXCR4 signaling in those lymph nodes.

FIGURE 1.

Expression of CXCR4 and SDF-1 in patients with oral SCC. A. Immunohistochemistry was used to examine the expression of SDF-1 and CXCR4 protein in the primary tumor and the metastatic lymph nodes in the same patients. Original magnification, ×400. a, SDF-1 in the primary tissue. b, CXCR4 in the primary tissue. c, SDF-1 in the metastatic lymph node. d, CXCR4 in the metastatic lymph node. B. Cause-specific Kaplan-Meier survival curve of patients with oral SCC with or without SDF-1.

FIGURE 1.

Expression of CXCR4 and SDF-1 in patients with oral SCC. A. Immunohistochemistry was used to examine the expression of SDF-1 and CXCR4 protein in the primary tumor and the metastatic lymph nodes in the same patients. Original magnification, ×400. a, SDF-1 in the primary tissue. b, CXCR4 in the primary tissue. c, SDF-1 in the metastatic lymph node. d, CXCR4 in the metastatic lymph node. B. Cause-specific Kaplan-Meier survival curve of patients with oral SCC with or without SDF-1.

Close modal
Table 1.

Expression of SDF-1 and CXCR4 in Primary Tumor and Metastatic Lymph Node

No.Age/GenderSDF-1
CXCR4
Survival Time (Mo/Status)
PrimaryMeta LNPrimaryMeta LN
68/M − 36/Dead 
76/M − − 16/Dead 
73/F − − 8/Dead 
66/M − − − 60/Alive 
70/F − 11/Dead 
64/M − − − 7/Dead 
78/M − − 8/Dead 
90/F − 8/Dead 
66/F − 29/Alive 
10 71/F − − − 60/Alive 
11 74/M − − 30/Dead 
12 61/M − − − − 52/Alive 
13 73/M − − − 8/Dead 
14 71/M − − − 11/Dead 
15 76/M 60/Alive 
16 64/M − 24/Dead 
17 83/F − − 60/Alive 
18 76/F − − − 60/Alive 
19 59/M − − 60/Alive 
20 76/M − 60/Alive 
21 67/F − − 22/Dead 
22 74/M − − − 60/Alive 
23 74/F − 58/Alive 
24 75/M 15/Dead 
25 73/F − − − 5/Dead 
26 80/M − 11/Dead 
27 62/F − 22/Dead 
28 71/M − − − 43/Alive 
29 70/M − − 60/Alive 
30 59/M − − 36/Alive 
Positive Number  16* 21 23  
No.Age/GenderSDF-1
CXCR4
Survival Time (Mo/Status)
PrimaryMeta LNPrimaryMeta LN
68/M − 36/Dead 
76/M − − 16/Dead 
73/F − − 8/Dead 
66/M − − − 60/Alive 
70/F − 11/Dead 
64/M − − − 7/Dead 
78/M − − 8/Dead 
90/F − 8/Dead 
66/F − 29/Alive 
10 71/F − − − 60/Alive 
11 74/M − − 30/Dead 
12 61/M − − − − 52/Alive 
13 73/M − − − 8/Dead 
14 71/M − − − 11/Dead 
15 76/M 60/Alive 
16 64/M − 24/Dead 
17 83/F − − 60/Alive 
18 76/F − − − 60/Alive 
19 59/M − − 60/Alive 
20 76/M − 60/Alive 
21 67/F − − 22/Dead 
22 74/M − − − 60/Alive 
23 74/F − 58/Alive 
24 75/M 15/Dead 
25 73/F − − − 5/Dead 
26 80/M − 11/Dead 
27 62/F − 22/Dead 
28 71/M − − − 43/Alive 
29 70/M − − 60/Alive 
30 59/M − − 36/Alive 
Positive Number  16* 21 23  

NOTE: −, negative staining; +, positive staining.

Abbreviations: ND, not determined; LN, lymph node.

*

Significantly increased than primary tissues (P < 0.001, by Wilcoxon signed rank test).

FIGURE 2.

Functional expression of autocrine SDF-1/CXCR4 in oral SCC cells. B88 cells were stably transfected by an empty vector or pEB6-SDF-1, isolated mock-transfectant cells, or B88-SDF-1 cells, respectively. A. The expression of SDF-1 mRNA and GAPDH mRNA in mock-transfectant cells or B88-SDF-1 cells. The cells were logarithmically grown and were harvested; RT-PCR was then done. B. Production of SDF-1 protein in mock or B88-SDF-1 cells. Conditioned medium from the logarithmically grown cells were harvested, and ELISA was done. C. Autonomous activation of extracellular signal-regulated kinase 1/2 (ERK1/2) and Akt/PKB in B88-SDF-1 cells. The cells were cultivated under serum-starved conditions and were harvested at the indicated time points. Western blot analysis of the kinases was done using antibodies that recognized the phosphorylated form (left) or total form (right) of ERK1/2 (top) or Akt/PKB (bottom). Data are representative of three separate experiments with similar results.

FIGURE 2.

Functional expression of autocrine SDF-1/CXCR4 in oral SCC cells. B88 cells were stably transfected by an empty vector or pEB6-SDF-1, isolated mock-transfectant cells, or B88-SDF-1 cells, respectively. A. The expression of SDF-1 mRNA and GAPDH mRNA in mock-transfectant cells or B88-SDF-1 cells. The cells were logarithmically grown and were harvested; RT-PCR was then done. B. Production of SDF-1 protein in mock or B88-SDF-1 cells. Conditioned medium from the logarithmically grown cells were harvested, and ELISA was done. C. Autonomous activation of extracellular signal-regulated kinase 1/2 (ERK1/2) and Akt/PKB in B88-SDF-1 cells. The cells were cultivated under serum-starved conditions and were harvested at the indicated time points. Western blot analysis of the kinases was done using antibodies that recognized the phosphorylated form (left) or total form (right) of ERK1/2 (top) or Akt/PKB (bottom). Data are representative of three separate experiments with similar results.

Close modal

Association between Expression of SDF-1/CXCR4 and Survival

Next, we investigated the association between the expression of SDF-1/CXCR4 and the survival of patients with oral SCC. The 5-year survival rate was 25.0% for the SDF-1–positive group and 71.4% for the SDF-1–negative group; that is, a significantly poorer outcome was observed for the SDF-1–positive group (P = 0.023, Fig. 1B). Among the cases of mortality involving SDF-1–positive cancer cells in the lymph nodes, secondary lung metastases were identified in six patients by computed tomography, indicating that autocrine SDF-1/CXCR4 might be involved in the distant metastasis. In addition, we also found a significant association in terms of the 5-year survival rate between the CXCR4-negative group (71.4%) and the CXCR4-positive group (39.1%) in the primary tumor (P = 0.035, data not shown).

Phenotypic Switch Induced by the Overexpression of SDF-1 in Oral SCC Cells

Because autocrine SDF-1/CXCR4 might contribute to the distant metastasis and poor prognosis of oral SCC, we isolated a SDF-1 stable transfectant in the CXCR4 positive oral SCC cells, B88, which expressed high levels of CXCR4. By means of reverse transcription-PCR (RT-PCR) and ELISA, this transfectant (B88-SDF-1) was found to express high levels of SDF-1 mRNA and protein (Fig. 2A and B). Autonomous extracellular signal-regulated kinase 1/2 and Akt/PKB activation were observed in B88-SDF-1 cells, indicating that this transfectant had acquired an autocrine SDF-1/CXCR4 system (Fig. 2C). We examined the expression of CXCR4 protein on the cell surface in the mock and B88-SDF-1 cells by use of fluorescence-activated cell sorting. However, we could not detect the significant change of CXCR4 expression in our assay system (data not shown). Although the overexpression of SDF-1 in B88 did not alter the anchorage-dependent growth of the cells (Fig. 5A), this transfectant significantly acquired chemokinetic response (Fig. 3A) and enhanced cell motility (Fig. 3B). Moreover, B88-SDF-1 cells also acquired the anchorage-independent growth potential compared with that of mock cells (Fig. 3C and D). We did the same experiment using an oral SCC cell line, HNt, in which the expression of CXCR4 is 7.5-fold lower than that in B88 cells (13). However, HNt-SDF-1 cells did exhibit slight, but not significant, phenotypic changes in vitro and in vivo, probably due to the reduced expression of CXCR4, compared with that of B88 cells (data not shown).

FIGURE 3.

Phenotypic switch by the autocrine SDF-1/CXCR4 system in oral SCC cells. A. Enhanced migration of B88-SDF-1 cells (black columns) compared with that of mock transfectant cells (white columns). Cells were seeded at 5 × 104 per well. After staining, the cells that plugged the pores and the cells attached to the lower surface of the membrane were counted in 10 fields under high-power magnification (×400). Bars, SD of triplicate samples. B. Enhanced wound healing associated with the B88-SDF-1 cells (middle) compared with that of mock (top) transfectants. After a generation of a linear wound on the confluent monolayers, cells were photographed at the same location on a grid 48 h later. The wound areas in each transfectant were calculated by NIH image (bottom). *, P = 0.0043. C. Acquisition of colony formation in soft agar of the B88-SDF-1 (right) transfectants. Top and bottom, cell colony formation at days 14 and 28, respectively. Magnification, ×400. D. The number of mock-transfectant (white columns) and B88-SDF-1 (black columns) colonies that formed. Columns, mean of triplicate samples; bars, SD. Data are representative of two separate experiments with similar results. *, P < 0.001 (statistically significant by one-way ANOVA).

FIGURE 3.

Phenotypic switch by the autocrine SDF-1/CXCR4 system in oral SCC cells. A. Enhanced migration of B88-SDF-1 cells (black columns) compared with that of mock transfectant cells (white columns). Cells were seeded at 5 × 104 per well. After staining, the cells that plugged the pores and the cells attached to the lower surface of the membrane were counted in 10 fields under high-power magnification (×400). Bars, SD of triplicate samples. B. Enhanced wound healing associated with the B88-SDF-1 cells (middle) compared with that of mock (top) transfectants. After a generation of a linear wound on the confluent monolayers, cells were photographed at the same location on a grid 48 h later. The wound areas in each transfectant were calculated by NIH image (bottom). *, P = 0.0043. C. Acquisition of colony formation in soft agar of the B88-SDF-1 (right) transfectants. Top and bottom, cell colony formation at days 14 and 28, respectively. Magnification, ×400. D. The number of mock-transfectant (white columns) and B88-SDF-1 (black columns) colonies that formed. Columns, mean of triplicate samples; bars, SD. Data are representative of two separate experiments with similar results. *, P < 0.001 (statistically significant by one-way ANOVA).

Close modal

Metastatic Potential in B88-SDF-1 Cells In vivo

Mock and B88-SDF-1 cells were orthotopically inoculated into the masseter muscle of nude mice (29). The size of the primary tumor did not differ between the two groups (data not shown). Although both mock and B88-SDF-1 cells histopathologically metastasized to the cervical lymph nodes, the weight of lymph nodes containing B88-SDF-1 cells was significantly heavier than that of lymph nodes containing mock cells (P = 0.0033; Fig. 4A). Moreover, we observed an increase in the number of metastatic lymph nodes and more aggressive metastatic foci in the lymph nodes containing B88-SDF-1 cells (Fig. 4B; Table 2). However, in the orthotopic inoculation experiment, no distant metastasis of either type of cell was detected at 30 days. Because nude mice with orthotopically inoculated mock or B88-SDF-1 cells were unable to survive for a long period of time due to the obstruction of the airway and/or the blockage of food intake, we next did i.v. inoculation using this transfectant. Consequently, the lungs with B88-SDF-1 cells were significantly heavier than those with mock cells (P = 0.0008; Fig. 4C). Moreover, numerous metastatic nodules were detected in the lungs in all of the mice inoculated with B88-SDF-1 cells (Fig. 4D, right). In contrast, lung metastasis was not observed in any of the mice inoculated with the mock transfectant (Fig. 4D, left). These results indicate that autocrine, rather than paracrine, SDF-1/CXCR4 signaling is necessary for the distant metastasis of oral SCC cells.

FIGURE 4.

Enhancement of lymph node metastasis and acquisition of lung metastasis in B88-SDF-1 cells. A and B. Cells were inoculated into the masseter muscle of nude mice (2 × 106), which were sacrificed at day 30. A. Mean weight of the bilateral cervical lymph nodes extracted from tumor-bearing nude mice. *, P = 0.0033 (one-way ANOVA). B. Metastatic cancer cells (brown zone) in the lymph nodes were immunostained by cytokeratin AE1/AE3 antibody. Data are representative of four lymph nodes, and the percentage staining area in all the lymph nodes is shown in Table 2. C and D. Cells (1 × 106) were inoculated into the blood vessels of nude mice, which were sacrificed at day 30. C. Mean weight of the lung in mock transfectant (white column) and B88-SDF-1 (black column) cells. **, P = 0.0008 (one-way ANOVA). D. Macroscopic (top) or histopathologic (bottom) lung features extracted from mock transfectant–inoculated (left) or B88-SDF-1–inoculated (right) nude mice.

FIGURE 4.

Enhancement of lymph node metastasis and acquisition of lung metastasis in B88-SDF-1 cells. A and B. Cells were inoculated into the masseter muscle of nude mice (2 × 106), which were sacrificed at day 30. A. Mean weight of the bilateral cervical lymph nodes extracted from tumor-bearing nude mice. *, P = 0.0033 (one-way ANOVA). B. Metastatic cancer cells (brown zone) in the lymph nodes were immunostained by cytokeratin AE1/AE3 antibody. Data are representative of four lymph nodes, and the percentage staining area in all the lymph nodes is shown in Table 2. C and D. Cells (1 × 106) were inoculated into the blood vessels of nude mice, which were sacrificed at day 30. C. Mean weight of the lung in mock transfectant (white column) and B88-SDF-1 (black column) cells. **, P = 0.0008 (one-way ANOVA). D. Macroscopic (top) or histopathologic (bottom) lung features extracted from mock transfectant–inoculated (left) or B88-SDF-1–inoculated (right) nude mice.

Close modal
Table 2.

Comparison of the Metastatic Lymph Node between Mock- and B88-SDF-1–Inoculated Nude Mice

Mouse No.% Tumor CellsNo. Metastatic LNs
B88-SDF-1 67.74* 
 54.94 
 51.59* 
 62.34* 
 64.04 
Mean ± SD  60.29 ± 6.692 1.6 ± 0.548 
    
Mock 31.25 
 45.01 
 31.78 
 32.30 
 ND 
Mean ± SD  35.08 ± 6.630 0.8 ± 0.447 
Mouse No.% Tumor CellsNo. Metastatic LNs
B88-SDF-1 67.74* 
 54.94 
 51.59* 
 62.34* 
 64.04 
Mean ± SD  60.29 ± 6.692 1.6 ± 0.548 
    
Mock 31.25 
 45.01 
 31.78 
 32.30 
 ND 
Mean ± SD  35.08 ± 6.630 0.8 ± 0.447 
*

Average of the percent tumor cells in the two lymph nodes.

P = 0.008, compared with mice inoculated with mock transfectant.

P = 0.035, compared with mice inoculated with mock transfectant.

Effect of a CXCR4 Antagonist, AMD3100, on the Metastasis of B88-SDF-1 Cells to the Lung

To assess the direct involvement of the autocrine SDF-1/CXCR4 system on the metastasis of B88-SDF-1 cells to the lungs, we first examined the effect of AMD3100, an antagonist for CXCR4, on the in vitro cell growth and motility. Although AMD3100 (1 μg/mL) did not influence the growth of the cells (Fig. 5A), AMD3100 significantly inhibited the enhanced motility of B88-SDF-1 cells both in wound assay (data not shown) and in transwell assay (Fig. 5A). Thus, we next treated the B88-SDF-1 tumor–bearing mice with AMD3100. Metastatic nodules in the lung were markedly inhibited by treatment with AMD3100, according to both macroscopic (data not shown) and histopathologic (Fig. 5B, top) data during the 50 days of observation. We also confirmed the presence of the metastatic cancer cells at the molecular level by use of RT-PCR extracted from the lungs at day 28 (Fig. 5C, middle). During the course of 25 cycles amplified by RT-PCR, no expression of human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was detected in three of four mice treated with AMD3100, but human GAPDH was detected in three of four mice treated by saline. Moreover, the treatment with AMD3100 significantly ameliorated the body weight loss of tumor-bearing nude mice (Fig. 5D). AMD3100 also significantly prolonged the survival rate of B88-SDF-1 tumor–bearing nude mice (Fig. 5E). It cannot be ruled out that the effects of AMD3100 were due to a combined inhibitory effect on both autocrine and paracrine SDF-1; however, these results indicated that the autocrine SDF-1/CXCR4 system were indispensable for the lung metastases induced by B88-SDF-1 and that AMD3100 may be an effective inhibitor of this autocrine SDF-1/CXCR4 system both in terms of distant metastasis and on the tumor-induced cachexia.

FIGURE 5.

Effect of AMD3100 on B88-SDF-1 cells in vitro and in vivo. The growth and motility of B88-SDF-1 cells with or without AMD3100 were examined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (A) and transwell assay (B), respectively. *, P < 0.0001. C. Molecular pathologic features of the lung extracted from control-treated (left) or AMD3100-treated (right) nude mice. Cells (1 × 106) were inoculated into the blood vessel of nude mice, which were sacrificed at day 30. Representative H&E staining of the lung (top) from control-treated (left) or AMD3100-treated (right) nude mice. RT-PCR was done by use of a pair of human (middle) or mouse (bottom) specific primers for GAPDH. D. Body weight of the control-treated (n = 8) and AMD3100-treated (n = 8) nude mice. The body weights of the mice were measured at days 0, 2, 5, 9, 12, 16, and 20. Points, mean; bars, SD. *, P < 0.05; †, P < 0.0001 (statistically significant by one-way ANOVA). E. Cause-specific Kaplan-Meier survival curve with B88-SDF-1–inoculated nude mice treated with or without AMD3100. The mice were observed at day 50. §, P = 0.0282 (statistically significant by a log-rank test).

FIGURE 5.

Effect of AMD3100 on B88-SDF-1 cells in vitro and in vivo. The growth and motility of B88-SDF-1 cells with or without AMD3100 were examined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (A) and transwell assay (B), respectively. *, P < 0.0001. C. Molecular pathologic features of the lung extracted from control-treated (left) or AMD3100-treated (right) nude mice. Cells (1 × 106) were inoculated into the blood vessel of nude mice, which were sacrificed at day 30. Representative H&E staining of the lung (top) from control-treated (left) or AMD3100-treated (right) nude mice. RT-PCR was done by use of a pair of human (middle) or mouse (bottom) specific primers for GAPDH. D. Body weight of the control-treated (n = 8) and AMD3100-treated (n = 8) nude mice. The body weights of the mice were measured at days 0, 2, 5, 9, 12, 16, and 20. Points, mean; bars, SD. *, P < 0.05; †, P < 0.0001 (statistically significant by one-way ANOVA). E. Cause-specific Kaplan-Meier survival curve with B88-SDF-1–inoculated nude mice treated with or without AMD3100. The mice were observed at day 50. §, P = 0.0282 (statistically significant by a log-rank test).

Close modal

In this study, we investigated the role of the autocrine SDF-1/CXCR4 system, focusing on the distant metastasis of oral SCC cells. The findings obtained from the present series of experiments were as follows. First, the number of SDF-1–positive metastatic lymph nodes was significantly higher than that of SDF-1–positive primary oral SCCs, which was associated with a poor survival rate among those of the former group. Second, the overexpression of SDF-1 in B88 cells conferred enhanced cell motility and anchorage-independent growth potential. Third, the transfectant acquired aggressive metastatic potential, including metastasis to the lymph nodes and lungs. Fourth, AMD3100, a CXCR4 antagonist, significantly inhibited the metastasis to the lungs of this transfectant as well as reducing tumor-induced cachexia. These results suggested that, in cases of oral SCC, the paracrine SDF-1/CXCR4 system potentiates lymph node metastasis, but the acquisition of an autocrine loop might be required for distant metastasis.

The clinical results showed a significantly higher number of SDF-1–positive cases among the metastatic lymph nodes than among the primary oral SCCs, which was associated with the poor survival rate among those of the former group. Unlike other CXC chemokines, SDF-1 is basally expressed in numerous tissues and in particular in mesenchymal cells (30, 31). The promoter of sdf1 contains several CpG islands, a transcription factor–binding motif commonly observed in housekeeping genes (30). Moreover, binding motifs for the transcription factors nuclear factor-κB and activator protein-1, which are commonly observed in other CXC chemokines, have not yet been found in the 19-kb sdf1 genomic clone (30). Therefore, it has generally been assumed that SDF-1 is a constitutively expressed chemokine; however, recent reports have suggested that SDF-1 expression is potently regulated by hypoxia-inducible factor-1 (32, 33). In addition, Staller et al. (34) showed that hypoxia-inducible factor positively regulates CXCR4 expression under hypoxic conditions, but the von Hippel-Lindau tumor suppressor protein pVHL negatively regulates CXCR4 expression, due to the capacity of the latter to target hypoxia-inducible factor for degradation under normoxic conditions. Because cancer cells that have escaped from a primary site are generally exposed to hypoxia due to a poor blood supply, not only SDF-1 but also CXCR4 expression in oral SCC might be regulated by certain tumor environmental factors such as hypoxia-inducible factor-1 at the metastatic lymph node, resulting in the acquisition of the autocrine loop of the SDF-1/CXCR4 system.

Oral SCCs are characterized by a high degree of local invasiveness and a high rate of metastases to the cervical lymph nodes; however, a low rate of metastases to distant organs is observed. Patients with distant metastases had high incidence of lymph node metastases (21). Moreover, we have previously reported that oral SCC cells easily intravasated into the circulation at early stage, despite the establishment of distant metastasis (35). These results indicating that distant metastasis in oral SCC is commonly arisen from the cancer cells which flow in from the metastatic lymph node, and that the failure of distant metastasis is due to a weak potential of extravasation or ectopic growth in oral SCC cells. In the present study, the autocrine SDF-1/CXCR4 system was found to enhance the cell migration and anchorage-independent (ectopic) growth potential in vitro. Furthermore, Iikura et al. (36) reported that the activation of the SDF-1/CXCR4 system caused strong transendothelial migration (extravasation) in basophils. These observations suggest that the autocrine SDF-1/CXCR4 system might induce extravasation or the ectopic growth potential of oral SCC cells, which are considered indispensable for the establishment of distant metastasis. During these processes, oral SCC cells are also exposed to disconnected paracrine SDF-1 gradient; however, constitutive autocrine signaling might be critical for the establishment of distant metastasis. Collectively, it is assumed that the following steps are required for the distant metastasis of oral SCC cells via SDF-1/CXCR4 system. First, CXCR4-related oral SCC cells metastasize to cervical lymph nodes via paracrine SDF-1 gradient, and acquire SDF-1 expression in the lymph node. Second, the cancer cells flow out from the lymph node, and enter superior vena cava and right atrial of the heart, which are pushed out from right ventral into the lungs, and attach to the alveolar capillary wall. Third, only the cells exhibiting SDF-1/CXCR4 autocrine loop could dominantly extravasate and grow into the alveolar space of the lungs.

Muller et al. (3) showed that paracrine SDF-1 produced by target organs such as the lung, liver, and bone marrow attracts CXCR4-expressing breast cancer cells. However, in the present study, oral SCC cells with an autocrine SDF-1/CXCR4 loop acquired aggressive metastatic potential, including that to the lymph nodes and lungs; however, cells exhibiting a paracrine loop did not metastasize to any distant organs, not even when introduced by i.v. inoculation. These observations suggest that paracrine SDF-1/CXCR4 signaling is insufficient for achieving distant metastasis in the case of oral SCC. Recent reports have shown that SDF-1 transactivates HER2-neu (37), which was shown to be amplified in 30% of breast cancers (38). Moreover, Li et al. (39) showed that HER2 enhances the expression of CXCR4, which is required for HER2-mediated invasion in vitro and lung metastasis in vivo. However, in the case of oral SCC, HER2 overexpression has been reported to be very rare; that is, rates of 3.6% (40), and even 0%, have been reported (41). Thus, breast cancer cells might use both paracrine SDF-1/CXCR4 and SDF-1/HER2 costimulatory pathways for the establishment of distant metastasis. In contrast to breast cancer, oral SCC might require strong constitutive SDF-1/CXCR4 autocrine signaling for the establishment of distant metastasis. Indeed, the autocrine mechanism of the SDF-1/CXCR system has been reported in studies of prostate cancer cells (42), thyroid cancer cells (9), osteosarcoma cells (43), and Kaposi's sarcoma cells (44), all of which are aggressive and highly distant metastatic tumors. These results indicate that the acquisition of the autocrine SDF-1/CXCR4 system might be a critical step in the acquisition of a more aggressive and metastatic phenotype among neoplastic cells. Thus, the expression of CXCR4 is an indicator of lymph node metastasis, and concomitant expression of SDF-1 might be a predictive marker of distant metastatic potential in oral SCC.

AMD3100 was first identified as a bicyclam derivative with potent activity against HIV infection (45). Subsequent studies clarified that AMD3100 selectively antagonized CXCR4, which acts as a coreceptor for HIV (46-48). Recent investigations have shown that AMD3100 also inhibits the invasion and metastasis of cancers expressing CXCR4, both in vitro and in vivo (49-51). These investigations exhibited the usefulness of AMD3100 as a tool for the evaluation of the physiologic and pathologic importance of SDF-1/CXCR4 interactions. In the present study, AMD3100 significantly inhibited the lung metastasis of B88-SDF-1 cells. Moreover, it ameliorated body weight loss and improved the survival rate of the tumor-bearing mice. These findings indicate that AMD3100 might also ameliorate tumor-induced cachexia in affected patients. In the future, AMD3100 might be used as an effective chemotherapeutic agent for patients with CXCR4-related oral SCC, and may also be a useful palliative drug for patients with advanced oral SCC.

Patients and Immunohistochemistry

Thirty patients with lymph node metastasis of advanced oral SCC classified stage IV (by International Union Against Cancer) were included in this study. Biopsy specimens and metastatic lymph nodes were obtained from 1990 to 1999 at the Second Department of Oral and Maxillofacial Surgery, Tokushima University Dental Hospital. All the patients selected in this study were primary treated by radiation, chemotherapy, and immunotherapy, as described previously (52). Immunohistochemistry was done with anti-CXCR4 monoclonal antibody (12G5; BioSource International) or anti–SDF-1α monoclonal antibody (R&D Systems, Inc.) as described previously (13). We defined as immunohistochemistry positive those cases in which over 25% of the cancer cells were stained for CXCR4 or SDF-1. The metastatic area in the lymph nodes was counted by the NIH image 1.63 after staining by the cytokeratin AE1/AE3 antibody (Dako Corporation).

Mice and In vivo Study

BALB/c nude mice were purchased from CLEA Japan. The mice were maintained under pathogen-free conditions and were handled in accordance with the Guidelines for Animal Experimentation of Tokushima University. The experiments were initiated when the mice were 8 weeks of age and were done as described previously (29). The presence or absence of lymph node and distant metastasis was confirmed by the H&E staining. In the experimental chemotherapy, every mouse was s.c. treated by an AMD3100 (2.5 mg/kg; Sigma) or same volume of saline at 48 h after the i.v. inoculation of the cells, as described previously (49). In some experiments, total RNA was prepared from lung using TRIzol (Invitrogen), and RT-PCR was done using a common upper primer, GAPDH-UP, CATCACCATCTTCCAGGAGCGA; and human and mouse specific lower primers, hGAPDH-DN, GGCATTGCTGATGATCTTGAGGCT, and mGAPDH-DN, TGCATTGCTGACAATCTTGAGTGA, respectively.

Cells and Cell Culture

B88 cells and HNt cells were maintained in DMEM supplemented with 10% FCS, 100 μg/mL streptomycin, and 100 units/mL penicillin in a humidified atmosphere of 95% air and 5% CO2 at 37°C.

Construction of a Mammalian Expression Vector, pEB6-SDF-1

EBV-based vector, pEB6-CAG-MCS, is an extrachromosomal vector carrying a replicational origin, oriP and a replication initiation factor (EBNA-1), which are sufficient for autonomous replication (53). The human SDF-1 cDNA fragment was amplified from placenta cDNA using a pair of specific primer, SDF-1UP, CCGCTCGAGCGCGCCATGAACGCCAAGGTCGTG, and SDF-1DN, CGGAATTCCATCTTGAACCTCTTGTTTAAAGC. This fragment, containing the human SDF-1 open reading frame, XhoI site at 5′, end and EcoRI site at 3′ end, was ligated to the cloning site of pEB6-CAG-MCS. The integrity and fidelity of the cloned sequences were confirmed by use of ABI 3100 sequencer (Applied Biosystems).

Stable Transfection

After seeding cells (5 × 106 per dish), the cells were transfected with 5 μg of pEB6-SDF-1 or pEB6-CAG-MCS, using Superfect (Qiagen). Forty-eight hours later, the cells were switched to a selective medium containing geneticin (400 μg/mL G418; Invitrogen Corp.). All of the G418 resistant clones were collected after 14 days of cultivation in the selective medium.

Enzyme-Linked Immunosorbent Assay

ELISA for SDF-1α was done by Quantikine kit (R&D) as described previously (10).

Western Blotting

Western blotting was done as described previously (10). The nitrocellulose membrane (Amersham Pharmacia Biotech) was incubated with primary antibodies against phosphorylated- or total- extracellular signal-regulated kinase 1/2 and Akt/PKB (Cell Signaling Technology), followed by horseradish peroxidase–conjugated secondary antibodies. Detection was then done using an enhanced chemiluminescence kit (Amersham Pharmacia Biotech).

Wound Assay

At 24 h after seeding the cells, a linear wound was generated on the confluent monolayers by scraping with a pipette chip with our without 1 μg/mL of AMD3100. Unattached cells were washed off with agitation. Cells were photographed at the same point on a grid at 48 h later. Each line was plated and wounded in triplicate.

In vitro Cell Migration Assay

The in vitro migration of oral SCC cells was evaluated using the Transwell (Corning) as described previously (10). The plugged cells in the pore or the cells attached to the lower surface of the membrane were counted in 10 fields at high-power view (×400) by a third person without any knowledge of the treatments. In some experiments, 1 μg/mL of AMD3100 was co-incubated with the cells, which were seeded on the upper chamber.

Soft Agar Assay

Transfectants (B88-SDF1 or B88-EB6) were seeded at a density of 1 × 104, 1 × 105, or 1 × 106 per well in six-well plates in 2 mL of 0.6% agar (Wako) supplemented with DMEM in the presence of 10% FCS. After 14 and 28 days, colonies containing >20 cells were counted.

Statistical Analysis

Statistical differences between the means for the different groups were evaluated with StatView 4.5 (Abacus Concepts) using one-way ANOVA, with the level of significance at P < 0.05. The cause-specific survival rates were calculated by the Kaplan-Meier method and compared using the log-rank test. The significance level was set at 5% for each analysis. All experiments were repeated twice to thrice, and similar results were obtained.

Grant support: Ministry of Education, Science and Culture of Japan grant 16791242.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: D. Uchida and T. Onoue contributed equally to this work.

We thank Drs. Fumie Omotehara, Naozumi Ishimaru, and Yoshio Hayashi (Department of Pathology, Tokushima University School of Dentistry) for their valuable histopathologic advice.

1
Rossi D, Zlotnik A. The biology of chemokines and their receptors.
Annu Rev Immunol
2000
;
18
:
217
–42.
2
Zlotnik A, Yoshie O. Chemokine a new classification system and their role in immunity.
Immunity
2000
;
12
:
121
–7.
3
Muller A, Homey B, Soto H, et al. Involvement of chemokine receptors in breast cancer metastasis.
Nature
2001
;
410
:
50
–6.
4
Scotton CJ, Wilson JL, Milliken D, Stamp G, Balkwill FR. Epithelial cancer cell migration: a role for chemokine receptors?
Cancer Res
2001
;
61
:
4961
–5.
5
Taichman RS, Cooper C, Keller ET, Pienta KJ, Taichman NS, McCauley LK. Use of the stromal cell-derived factor-1/CXCR4 pathway in prostate cancer metastasis to bone.
Cancer Res
2002
;
62
:
1832
–7.
6
Schrader AJ, Lechner O, Templin M, et al. CXCR4/CXCL12 expression and signalling in kidney cancer.
Br J Cancer
2002
;
86
:
1250
–6.
7
Zhou Y, Larsen PH, Hao C, Yong VW. CXCR4 is a major chemokine receptor on glioma cells and mediates their survival.
J Biol Chem
2002
;
277
:
49481
–7.
8
Kijima T, Maulik G, Ma PC, et al. Regulation of cellular proliferation, cytoskeletal function, and signal transduction through CXCR4 and c-Kit in small cell lung cancer cells.
Cancer Res
2002
;
62
:
6304
–11.
9
Hwang JH, Hwang JH, Chung HK, et al. CXC chemokine receptor 4 expression and function in human anaplastic thyroid cancer cells.
J Clin Endocrinol Metab
2003
;
88
:
408
–16.
10
Uchida D, Begum NM, Almofti A, et al. Possible role of stromal-cell-derived factor-1/CXCR4 signaling on lymph node metastasis of oral squamous cell carcinoma.
Exp Cell Res
2003
;
290
:
289
–302.
11
Uchida D, Begum NM, Tomizuka Y, et al. Acquisition of lymph node, but not distant metastatic potentials, by the overexpression of CXCR4 in human oral squamous cell carcinoma.
Lab Invest
2004
;
84
:
1538
–46.
12
Onoue T, Uchida D, Begum NM, Tomizuka Y, Yoshida H, Sato M. Epithelial-mesenchymal transition induced by the stromal cell-derived factor-1/CXCR4 system in oral squamous cell carcinoma cells.
Int J Oncol
2006
;
29
:
1133
–8.
13
Almofti A, Uchida D, Begum NM, et al. The clinicopathological significance of the expression of CXCR4 protein in oral squamous cell carcinoma.
Int J Oncol
2004
;
25
:
65
–71.
14
Ishikawa T, Nakashiro K, Hara S, et al. CXCR4 expression is associated with lymph-node metastasis of oral squamous cell carcinoma.
Int J Oncol
2006
;
28
:
61
–6.
15
Delilbasi CB, Okura M, Iida S, Kogo M. Investigation of CXCR4 in squamous cell carcinoma of the tongue.
Oral Oncol
2004
;
40
:
154
–7.
16
Hu J, Deng X, Bian X, et al. The expression of functional chemokine receptor CXCR4 is associated with the metastatic potential of human nasopharyngeal carcinoma.
Clin Cancer Res
2005
;
11
:
4658
–65.
17
Kaifi JT, Yekebas EF, Schurr P, et al. Tumor-cell homing to lymph nodes and bone marrow and CXCR4 expression in esophageal cancer.
J Natl Cancer Inst
2005
;
97
:
1840
–7.
18
Zlotnik A. Chemokines in neoplastic progression.
Semin Cancer Biol
2004
;
14
:
181
–5.
19
Kowalski LP, Carvalho AL, Martins Priante AV, Magrin J. Predictive factors for distant metastasis from oral and oropharyngeal squamous cell carcinoma.
Oral Oncol
2005
;
41
:
534
–41.
20
Brouwer J, de Bree R, Hoekstra OS, et al. Screening for distant metastases in patients with head and neck cancer: is chest computed tomography sufficient?
Laryngoscope
2005
;
115
:
1813
–7.
21
Alvi A, Johnson JT. Development of distant metastasis after treatment of advanced-stage head and neck cancer.
Head Neck
1997
;
19
:
500
–5.
22
Nicolson GL. Tumor and host molecules important in the organ preference of metastasis.
Semin Cancer Biol
1991
;
2
:
143
–54.
23
Nicolson GL. Paracrine and autocrine growth mechanisms in tumor metastasis to specific sites with particular emphasis on brain and lung metastasis.
Cancer Metastasis Rev
1993
;
12
:
325
–43.
24
Jechlinger M, Sommer A, Moriggl R, et al. Autocrine PDGFR signaling promotes mammary cancer metastasis.
J Clin Invest
2006
;
116
:
1561
–70.
25
Tsutsumi S, Yanagawa T, Shimura T, Kuwano H, Raz A. Autocrine motility factor signaling enhances pancreatic cancer metastasis.
Clin Cancer Res
2004
;
10
:
7775
–84.
26
Otsuka T, Takayama H, Sharp R, et al. c-Met autocrine activation induces development of malignant melanoma and acquisition of the metastatic phenotype.
Cancer Res
1998
;
58
:
5157
–67.
27
Oft M, Heider KH, Beug H. TGFβ signaling is necessary for carcinoma cell invasiveness and metastasis.
Curr Biol
1998
;
8
:
1243
–52.
28
Sugiyama K, Yonemura Y, Miyazaki I. Immunohistochemical study of epidermal growth factor and epidermal growth factor receptor in gastric carcinoma.
Cancer
1989
;
63
:
1557
–61.
29
Kawamata H, Nakashiro K, Uchida D, Harada K, Yoshida H, Sato M. Possible contribution of active MMP2 to lymph-node metastasis and secreted cathepsin L to bone invasion of newly established human oral-squamous-cancer cell lines.
Int J Cancer
1997
;
70
:
120
–7.
30
Shirozu M, Nakano T, Inazawa J, et al. Structure and chromosomal localization of the human stromal cell-derived factor 1 (SDF1) gene.
Genomics
1995
;
28
:
495
–500.
31
Tashiro K, Tada H, Heilker R, Shirozu M, Nakano T, Honjo T. Signal sequence trap: a cloning strategy for secreted proteins and type I membrane proteins.
Science
1993
;
261
:
600
–3.
32
Hitchon C, Wong K, Ma G, Reed J, Lyttle D, El-Gabalawy H. Hypoxia-induced production of stromal cell-derived factor 1 (CXCL12) and vascular endothelial growth factor by synovial fibroblasts.
Arthritis Rheum
2002
;
46
:
2587
–97.
33
Ceradini DJ, Kulkarni AR, Callaghan MJ, et al. Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1.
Nat Med
2004
;
10
:
858
–64.
34
Staller P, Sulitkova J, Lisztwan J, Moch H, Oakeley EJ, Krek W. Chemokine receptor CXCR4 downregulated by von Hippel-Lindau tumour suppressor pVHL.
Nature
2003
;
425
:
307
–11.
35
Kawamata H, Uchida D, Nakashiro K, et al. Haematogenous cytokeratin 20 mRNA as a predictive marker for recurrence in oral cancer patients.
Br J Cancer
1999
;
80
:
448
–52.
36
Iikura M, Ebisawa M, Yamaguchi M, et al. Transendothelial migration of human basophils.
J Immunol
2004
;
173
:
5189
–95.
37
Cabioglu N, Summy J, Miller C, et al. CXCL-12/stromal cell-derived factor-1α transactivates HER2-neu in breast cancer cells by a novel pathway involving Src kinase activation.
Cancer Res
2005
;
65
:
6493
–7.
38
Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene.
Science
1987
;
235
:
177
–82.
39
Li YM, Pan Y, Wei Y, et al. Upregulation of CXCR4 is essential for HER2-mediated tumor metastasis.
Cancer Cell
2004
;
6
:
459
–69.
40
Freier K, Joos S, Flechtenmacher C, et al. Tissue microarray analysis reveals site-specific prevalence of oncogene amplifications in head and neck squamous cell carcinoma.
Cancer Res
2003
;
63
:
1179
–82.
41
Riviere A, Becker J, Loning T. Comparative investigation of c-erbB2/neu expression in head and neck tumors and mammary cancer.
Cancer
1991
;
67
:
2142
–9.
42
Sun YX, Wang J, Shelburne CE, et al. Expression of CXCR4 and CXCL12 (SDF-1) in human prostate cancers (PCa) in vivo.
J Cell Biochem
2003
;
89
:
462
–73.
43
Perissinotto E, Cavalloni G, Leone F, et al. Involvement of chemokine receptor 4/stromal cell-derived factor 1 system during osteosarcoma tumor progression.
Clin Cancer Res
2005
;
11
:
490
–7.
44
Wang JF, Liu ZY, Anand AR, et al. α-chemokine-mediated signal transduction in human Kaposi's sarcoma spindle cells.
Biochim Biophys Acta
2004
;
1691
:
129
–39.
45
De Clercq E, Yamamoto N, Pauwels R, et al. Potent and selective inhibition of human immunodeficiency virus (HIV)-1 and HIV-2 replication by a class of bicyclams interacting with a viral uncoating event.
Proc Natl Acad Sci U S A
1992
;
89
:
5286
–90.
46
Schols D, Struyf S, Van Damme J, Este JA, Henson G, De Clercq E. Inhibition of T-tropic HIV strains by selective antagonization of the chemokine receptor CXCR4.
J Exp Med
1997
;
186
:
1383
–8.
47
Donzella GA, Schols D, Lin SW, et al. AMD3100, a small molecule inhibitor of HIV-1 entry via the CXCR4 co-receptor.
Nat Med
1998
;
4
:
72
–7.
48
Feng Y, Broder CC, Kennedy PE, Berger EA. HIV-1 entry cofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor.
Science
1996
;
272
:
872
–7.
49
Rubin JB, Kung AL, Klein RS, et al. A small-molecule antagonist of CXCR4 inhibits intracranial growth of primary brain tumors.
Proc Natl Acad Sci U S A
2003
;
100
:
13513
–8.
50
Scotton CJ, Wilson JL, Scott K, et al. Multiple actions of the chemokine CXCL12 on epithelial tumor cells in human ovarian cancer.
Cancer Res
2002
;
62
:
5930
–8.
51
Smith MC, Luker KE, Garbow JR, et al. CXCR4 regulates growth of both primary and metastatic breast cancer.
Cancer Res
2004
;
64
:
8604
–12.
52
Sato M, Harada K, Yoshida H, et al. Therapy for oral squamous cell carcinoma by tegafur and streptococcal agent OK-432 in combination with radiotherapy: association of the therapeutic effect with differentiation and apoptosis in the cancer cells.
Apoptosis
1997
;
2
:
227
–38.
53
Tanaka J, Miwa Y, Miyoshi K, Ueno A, Inoue H. Construction of Epstein-Barr virus-based expression vector containing mini-oriP.
Biochem Biophys Res Commun
1999
;
264
:
938
–43.