Approximately 30% of patients with renal cell carcinoma (RCC) develop bone metastasis, which is characterized by extensive osteolysis leading to severe bone pain and pathologic fracture. Although the mechanism of RCC-induced osteolysis is unknown, studies of bone metastasis have shown that tumor-induced changes in bone remodeling are likely mediated by alterations in the bone microenvironment. Here, we report the discovery of a novel osteoclast stimulatory factor secreted by RCC bone metastasis (RBM). Through microarray analysis, we found expression of the chemokine, macrophage inflammatory protein-1δ (MIP-1δ), to be increased in RBM versus patient-matched primary RCC tissues and confirmed this finding by quantitative reverse transcription-PCR (qRT-PCR) and ELISA (P < 0.05). Furthermore, MIP-1δ expression in RBM tissues was significantly (P < 0.001) higher than in human bone marrow, suggesting a potential alteration of the bone microenvironment. The receptors for MIP-1δ, CCR1 and CCR3, were expressed in both osteoclast precursors and mature, bone-resorbing osteoclasts as shown by qRT-PCR and Western analysis. In functional studies, MIP-1δ stimulated chemotaxis of two osteoclast precursor cell types: murine bone marrow mononuclear cells (BM-MNC) and RAW 264.7 cells. Furthermore, MIP-1δ treatment of murine calvaria caused increased bone resorption as determined by measurement of released calcium. Correspondingly, MIP-1δ significantly enhanced osteoclast formation and activity in response to RANKL in both BM-MNC and RAW 264.7 cells. Taken together, these data suggest that MIP-1δ expression is increased in RBM relative to RCC and bone marrow, and may promote RBM-induced osteolysis by stimulating the recruitment and differentiation of osteoclast precursors into mature osteoclasts. [Cancer Res 2008;68(5):1261–6]

Renal cell carcinoma (RCC) is the most common malignant tumor arising from the kidney, accounting for ∼30,000 new cases per year in the United States. In 30% of these cases, bone metastases will develop, causing marked morbidity in the form of severe bone pain, decreased mobility, neurologic compromise, and pathologic fracture (1). These complications result from the ability of RCC cells to disrupt the delicate balance of bone remodeling, leading to pathologic bone loss. Unfortunately, despite increased research, the mechanisms through which RCC cells alter the bone remodeling process remain unknown.

Chemokines are well-known for mediating inflammatory responses by stimulating the recruitment of leukocytes; however, recent evidence suggests that they may also play a role in mediating pathologic bone loss. The chemokine, interleukin-8, has been implicated as a mediator of osteoclast-mediated bone resorption in breast cancer bone metastasis (2, 3). In addition, studies suggest that macrophage inflammatory protein (MIP)-1α plays a role in mediating osteolysis associated with multiple myeloma. Bone marrow plasma from multiple myeloma patients show increased levels of MIP-1α relative to normal controls (4). Furthermore, MIP-1α stimulates osteoclast development in human bone marrow cultures (5).

MIP-1δ (CCL15), originally called leukotactin-1, was discovered in 1997 and exhibits 46% homology with MIP-1α at the amino acid level (6). Like MIP-1α, MIP-1δ belongs to the MIP family of CC-type chemokines, which currently consists of six members: MIP-1α, MIP-1β, MIP-1δ, MIP-1γ, MIP-3α, and MIP-3β. MIP proteins are mainly produced by leukocytes after exposure to inflammatory cytokines and play a major role in the recruitment of immune cells to sites of injury or infection. MIP-1δ exerts its biological effects through the chemokine receptors CCR1 and CCR3 (6). Although both MIP-1α and MIP-1δ use CCR1, MIP-1δ is a stronger CCR1 agonist. Similar to other MIP family members, MIP-1δ induces chemotaxis in immune cells. However, contrary to MIP-1α, the role of MIP-1δ in osteoclast differentiation has not been examined. Here, we present, for the first time, evidence that MIP-1δ levels are significantly elevated in RCC bone metastasis (RBM) relative to RCC and normal bone marrow. Furthermore, MIP-1δ stimulates recruitment of osteoclast precursors and enhances RANKL-induced osteoclast differentiation and activity. These data suggest that MIP-1δ is a novel osteoclast-stimulating factor, which may promote RBM-induced osteolysis.

Cell lines and culture conditions. RBM cell lines were maintained as described (7). RAW 264.7 cells were purchased from American Type Culture Collection and maintained according to conditions specified. Murine bone marrow mononuclear cells (BM-MNC) were isolated from 4- to 8-week-old mice by flushing the bone with 10 mL α-MEM containing 15% fetal bovine serum (FBS) + 1% penicillin/streptomycin (P/S) using a 26-gauge needle. Cells were washed, resuspended in α-MEM + 10% FBS + 1% P/S, and incubated overnight at 37°C. Nonadherent cells were collected, strained through 70-μm nylon mesh, and used in experimental assays as described.

Calvarial bone resorption assay. Neonatal mice (5 days) were euthanized, calvaria were removed, and incubated in α-MEM containing 1% bovine serum albumin + 1% P/S for 24 hours at 37°C. Calvaria were then washed with PBS and incubated in α-MEM containing respective factors. Media were replaced every 3 days, and the assay was conducted for a maximum of 7 days. Bone resorption was assessed by measuring calcium released into the medium using a calcium assay kit (Bioassay System).

ELISA. Total protein was extracted from tumor tissues in radioimmunoprecipitation assay buffer, conditioned media were collected from RBM cell lines, and bone marrow was collected from healthy donors at Johns Hopkins Hospital, observing institutional guidelines for specimen acquisition. MIP-1δ levels were analyzed using Quantikine MIP-1δ enzyme immunoassay kit (R&D Systems).

Migration assay. BM-MNC or RAW 264.7 cells suspended in complete medium were added to the upper chamber of transwell units with 8-μm pore size (Becton Dickinson). Inserts were then placed into the lower chambers of transwell units containing MIP-1δ in complete medium. Transwell units were incubated at 37°C for 3 hours, fixed in methanol, and cells on the upper side of the transwell membrane were removed with a cotton swab. Membranes were stained with 0.5% toluidine blue, and migrated cells were counted by light microscopy.

Osteoclast differentiation assay. BM-MNC or RAW 264.7 cells were incubated at 37°C for 3 days in complete medium supplemented with or without MIP-1δ alone or in the presence of MIP-1δ neutralizing antibody (R&D Systems). Medium was then removed, and complete medium with or without RANKL (50 ng/mL) was added. Half the medium was replaced every 2 days. After 10 days, the development of mature osteoclasts was determined by positive tartrate-resistant acidic phosphatase (TRAP) staining and the presence of ≥3 nuclei under light microscopy. To determine the fusion index, the number of nuclei per osteoclast was counted in 15 to 20 random fields at ×100 magnification.

Osteoclast activity assay. BM-MNC or RAW 264.7 cells were cultured in an Osteoassay human bone plate at 37°C for 3 days in complete medium with or without MIP-1δ alone or in the presence of MIP-1δ neutralizing antibody (R&D Systems). Medium was then removed, and complete medium with or without RANKL was added for 12 days. Half the medium was replaced every 2 to 3 days. Levels of released type I collagen helical peptide in supernatant samples were then measured by Metra Helical Peptide ELISA kit (Quidel) as a measure of osteoclast activity.

Quantitative reverse transcription-PCR. Total RNA was extracted using Trizol (Invitrogen), and cDNA was generated by reverse transcription. Twenty-five microliters of reactions contained 1× SYBR Green Reaction Mix (Bio-Rad), 1 μL cDNA, and 100 nm of each primer: MIP-1δ, 5′-CTCTCCTGCCTCATGCTTGT-3′ (sense) and 5′-GCACTCGCTGCTCGTTTCAA-3′ (antisense); CCR1, 5′-ACTCCACTCCATGCCAAAAGA-3′ (sense) and 5′-GCAAACACAGCATGGACAATG-3′ (antisense); CCR3, 5′-AAAAACACAAGGCCATCCGT-3′ (sense) and 5′-TCCATTTTCTCACCAGGAAG-3′ (antisense); and 36B4, 5′-GAAGGCTGTGGTGCTGATGG-3′ (sense) and 5′-CCCCTGGAGATTTTAGTGGT-3′ (antisense). Quantitative reverse transcription-PCR (qRT-PCR) variables were as follows: 1 cycle (95°C for 3 minutes) and 40 cycles [95°C for 30 s; 55.5°C (CCR1), 61.9°C (CCR3), and 65°C (MIP-1δ; 36B4) for 30 s; and 72°C for 45 s]. Amplification of 36B4 was used as an internal control. Relative expression among samples was calculated by the comparative CT method.

Western blotting. Total protein was extracted from cell lines using lysis buffer consisting of 15% glycerol, 5% SDS, and 250 mmol/L Tris-HCl (pH 6.7). Equal amounts of protein were resolved using 10% SDS-PAGE. Protein was transferred to enhanced chemiluminescence nitrocellulose membranes (Amersham) and probed with anti-CCR1 (Abcam), CCR3 (Epitomics), or β-actin (Amersham) antibody. Membranes were then incubated with horseradish peroxidase–conjugated antibody against rabbit or mouse IgG (Amersham), and binding was revealed by chemiluminescence (Amersham).

Expression of MIP-1δ in RBM. To identify genes with potential involvement in RBM growth and osteolysis, we compared gene expression between two human RBM tissues and patient-matched primary RCC tissues by oligonucleotide microarray (data not shown). Microarray analysis suggested that mRNA expression of the chemokine MIP-1δ was increased in RBM tissues relative to patient-matched primary RCC tissues. To confirm this finding, we performed qRT-PCR to directly compare the MIP-1δ mRNA expression levels in the RBM and RCC tissue samples used for microarray analysis. Corresponding with microarray data, MIP-1δ mRNA expression was elevated in RBM tissues (RBM4 and RBM29) relative to patient-matched RCC tissues (4P and 29P; Fig. 1A). Similarly, when expanding the analysis to include additional RBM and RCC tissue samples, MIP-1δ mRNA expression in RBM tissues was ∼3-fold higher on average compared with RCC tissues, although expression levels were variable. This finding was confirmed at the protein level where MIP-1δ protein levels were significantly (P < 0.05) increased in RBM tissues relative to RCC tissues (Fig. 1B). Furthermore, consistent with mRNA expression, MIP-1δ protein levels were significantly (P < 0.001) elevated in RBM tissues relative to bone marrow from disease-free donors, suggesting a potential alteration of the bone microenvironment in RBM patients (Fig. 1B). MIP-1δ mRNA and protein levels were also consistently expressed in six RBM cell lines by qRT-PCR (data not shown) and ELISA, respectively (Fig. 1C).

Figure 1.

Expression of MIP-1δ in human RBM. A and D, MIP-1δ mRNA expression was determined by qRT-PCR. B and C, MIP-1δ protein expression was determined by ELISA. Samples are (A–C) human bone marrow (BM), primary RCC tissue samples (RCC), RBM tissue samples (RBM), bone marrow samples from disease-free donors (BM), and conditioned medium from RBM cell lines; D, bone metastases from prostate cancer (PBM1), breast cancer (BBM), myeloma (MYBM), melanoma (MBM), thyroid cancer (TBM), and RCC (RBM). Data were obtained from experiments performed in triplicate. Columns, mean; bars, SE. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 1.

Expression of MIP-1δ in human RBM. A and D, MIP-1δ mRNA expression was determined by qRT-PCR. B and C, MIP-1δ protein expression was determined by ELISA. Samples are (A–C) human bone marrow (BM), primary RCC tissue samples (RCC), RBM tissue samples (RBM), bone marrow samples from disease-free donors (BM), and conditioned medium from RBM cell lines; D, bone metastases from prostate cancer (PBM1), breast cancer (BBM), myeloma (MYBM), melanoma (MBM), thyroid cancer (TBM), and RCC (RBM). Data were obtained from experiments performed in triplicate. Columns, mean; bars, SE. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

To determine whether expression of MIP-1δ was common to bone metastases in general, we examined MIP-1δ mRNA expression in RBM tissues versus nonrenal bone metastases by qRT-PCR. Interestingly, MIP-1δ mRNA expression in RBM tissues was significantly (P < 0.001) higher than in bone metastases from prostate, breast, myeloma, melanoma, and thyroid cancers (Fig. 1D). Furthermore, the levels of MIP-1δ mRNA expressed in nonrenal bone metastases were not significantly different from that detected in bone marrow. These data suggest that MIP-1δ is expressed in RBM tissues at levels significantly greater than those observed in RCC or bone marrow and is not common to bone metastasis from other cancer types.

MIP-1δ stimulates chemotaxis of osteoclast precursors. Previous studies showed that the biological effects of MIP-1δ are exerted through the chemokine receptors CCR1 and CCR3 (6). To assess whether secretion of MIP-1δ by RBM could elicit paracrine effects on osteoclasts, potentially contributing to RBM-induced osteolysis, we examined the expression of the chemokine receptors CCR1 and CCR3 in osteoclast precursors and mature osteoclasts from both murine bone marrow and RAW 264.7 cell models of osteoclast differentiation. Expression of CCR1 was detectable in osteoclast precursors and was elevated in mature osteoclasts by qRT-PCR and Western analysis (Fig. 2A and C). CCR3 expression was also elevated in mature osteoclasts relative to osteoclast precursors using BM-MNC; however, osteoclast precursors and mature osteoclasts derived from RAW 264.7 cells displayed equivalent levels of CCR3 (Fig. 2B and C). Taken together, these data show that MIP-1δ receptors CCR1 and CCR3 are expressed in both osteoclast precursors and mature osteoclasts, suggesting that MIP-1δ may have biological effects on each of these cell populations.

Figure 2.

CCR1 and CCR3 are expressed in osteoclast precursors and up-regulated in mature osteoclasts. A and B, CCR1 and CCR3 mRNA expression was determined by qRT-PCR. Data were obtained from experiments performed in triplicate. Columns, mean relative to other samples as calculated by the comparative CT method; bars, SE. *, P < 0.05; ***, P < 0.001. C, CCR1 and CCR3 protein expression was determined by Western analysis performed on equal amounts of protein from total cell lysates. OCL, osteoclast.

Figure 2.

CCR1 and CCR3 are expressed in osteoclast precursors and up-regulated in mature osteoclasts. A and B, CCR1 and CCR3 mRNA expression was determined by qRT-PCR. Data were obtained from experiments performed in triplicate. Columns, mean relative to other samples as calculated by the comparative CT method; bars, SE. *, P < 0.05; ***, P < 0.001. C, CCR1 and CCR3 protein expression was determined by Western analysis performed on equal amounts of protein from total cell lysates. OCL, osteoclast.

Close modal

To begin exploring the potential paracrine effect of RBM-produced MIP-1δ on bone cells, we examined the ability of MIP-1δ to induce the chemotactic recruitment of osteoclast precursors. When present solely in the lower chamber of transwell migration units, MIP-1δ stimulated the migration of BM-MNC (Fig. 3A) and RAW 264.7 (Fig. 3B) cells in a dose-dependent fashion. Increased migration was not due to an increase in random cell movement (chemokinesis) because stimulation of cell migration did not occur when MIP-1δ was present in equal concentrations in the top and bottom chambers of the transwell units (data not shown). Furthermore, migration was abrogated in the presence of MIP-1δ–neutralizing antibody. These data suggest that MIP-1δ secretion by RBM may promote the recruitment of osteoclast precursors.

Figure 3.

MIP-1δ induces chemotaxis of osteoclast precursors. Murine (A) BM-MNC and (B) RAW 264.7 cells were loaded into the top chamber of transwell filter units and allowed to migrate (3 h) toward MIP-1δ with or without neutralizing antibody (Ab) in the lower chamber. Nonmigrating cells were removed from the upper surface of the transwell filter, and migrating cells on the lower surface were stained and quantified. Data were obtained from three independent experiments having four replicates per condition. Columns, mean percentage of control (cell migration in the absence of MIP-1δ addition); bars, SE. Significant differences displayed are relative to control or as otherwise indicated. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 3.

MIP-1δ induces chemotaxis of osteoclast precursors. Murine (A) BM-MNC and (B) RAW 264.7 cells were loaded into the top chamber of transwell filter units and allowed to migrate (3 h) toward MIP-1δ with or without neutralizing antibody (Ab) in the lower chamber. Nonmigrating cells were removed from the upper surface of the transwell filter, and migrating cells on the lower surface were stained and quantified. Data were obtained from three independent experiments having four replicates per condition. Columns, mean percentage of control (cell migration in the absence of MIP-1δ addition); bars, SE. Significant differences displayed are relative to control or as otherwise indicated. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

MIP-1δ enhances osteoclast differentiation and bone resorption. To test the ability of MIP-1δ to induce bone resorption, we used the calvarial bone resorption assay, an ex vivo bone culture system in which known factors induce bone resorption primarily through the induction of osteoclastogenesis (8). Treatment of neonatal murine calvaria with MIP-1δ increased bone resorption relative to medium alone, as evidenced by calcium release (Fig. 4A), suggesting that MIP-1δ may enhance osteoclastogenesis. To test this possibility, we examined the effect of MIP-1δ on the differentiation of BM-MNC and RAW 264.7 osteoclast precursors. Although exposure to MIP-1δ alone was unable to stimulate osteoclast formation, MIP-1δ treatment potentiated RANKL-induced osteoclast formation in both BM-MNC (Fig. 4B) and RAW 264.7 (Fig. 4C) osteoclast precursor cells in a dose-dependent fashion, as evidenced by formation of TRAP-positive multinucleated cells. Furthermore, osteoclasts formed in BM-MNC and RAW 264.7 cultures treated with MIP-1δ had a significantly (P < 0.001) higher fusion index (13.4 and 14.2 nuclei per osteoclast, respectively) relative to those treated with RANKL alone (8.3 and 7.8 nuclei per osteoclast; Supplementary Fig. S1). Consistent with enhanced osteoclast formation, combined stimulation of osteoclast precursors with MIP-1δ and RANKL increased the degradation of human bone fragments in an Osteoassay human bone plate as evidenced by release of type I collagen helical peptide (Fig. 4D). The ability of MIP-1δ to enhance osteoclast differentiation and activity was further supported by the abrogation of these effects in the presence of MIP-1δ–neutralizing antibody (Fig. 4B–D). Thus, these studies suggest that MIP-1δ secretion by RBM may contribute to RBM-induced osteolysis by promoting the differentiation of osteoclast precursors into active osteoclasts.

Figure 4.

MIP-1δ enhances osteoclast differentiation and bone resorption in vitro. A, neonatal calvaria were treated with medium alone, tumor necrosis factor α (TNF-α) + M-CSF (positive control), or MIP-1δ for 5 d. Bone resorption was assessed by measurement of calcium released into the medium. B and C, murine (B) BM-MNC and (C) RAW 264.7 cells were treated with or without increasing concentrations of MIP-1δ in the presence or absence of neutralizing antibody for 3 d. Medium was then removed, and cells were incubated in fresh complete medium containing RANKL (50 ng/mL) for 10 d. Osteoclasts were identified by positive TRAP staining and the presence of ≥ 3 nuclei. D, BM-MNC were seeded on human bone fragments in an Osteoassay human bone plate and treated with or without MIP-1δ (0.1 pg/mL) in the presence or absence of neutralizing antibody for 3 d, followed by RANKL (50 ng/mL) for 10 d. Bone resorption was then assessed by measurement of type I collagen helical peptide fragments released into the medium over time by ELISA. Data were obtained from four replicates per condition in two independent experiments. Columns, mean; bars, SE. Significant differences displayed are relative to negative control (A), RANKL treatment alone (BD), or as otherwise indicated. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 4.

MIP-1δ enhances osteoclast differentiation and bone resorption in vitro. A, neonatal calvaria were treated with medium alone, tumor necrosis factor α (TNF-α) + M-CSF (positive control), or MIP-1δ for 5 d. Bone resorption was assessed by measurement of calcium released into the medium. B and C, murine (B) BM-MNC and (C) RAW 264.7 cells were treated with or without increasing concentrations of MIP-1δ in the presence or absence of neutralizing antibody for 3 d. Medium was then removed, and cells were incubated in fresh complete medium containing RANKL (50 ng/mL) for 10 d. Osteoclasts were identified by positive TRAP staining and the presence of ≥ 3 nuclei. D, BM-MNC were seeded on human bone fragments in an Osteoassay human bone plate and treated with or without MIP-1δ (0.1 pg/mL) in the presence or absence of neutralizing antibody for 3 d, followed by RANKL (50 ng/mL) for 10 d. Bone resorption was then assessed by measurement of type I collagen helical peptide fragments released into the medium over time by ELISA. Data were obtained from four replicates per condition in two independent experiments. Columns, mean; bars, SE. Significant differences displayed are relative to negative control (A), RANKL treatment alone (BD), or as otherwise indicated. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

Patients with RBM experience marked morbidity due to excessive bone destruction, yet the mechanisms responsible for this pathologic event remain unknown. Here, for the first time, we have identified MIP-1δ as a novel osteoclast stimulatory factor produced at elevated levels in human RBM tissues relative to primary RCC tissues and bone marrow from healthy adults. Strikingly, MIP-1δ not only stimulated chemotaxis of osteoclast precursors but also significantly (P ≤ 0.01) enhanced osteoclast differentiation and activity in response to RANKL. Taken together, these studies suggest that MIP-1δ secretion by RBM may contribute to RBM-induced osteolysis via recruitment and differentiation of osteoclast precursors into active osteoclasts.

Although our study clearly shows the ability of MIP-1δ to directly enhance osteoclast formation in combination with RANKL, there are additional means through which MIP-1δ may influence RBM-induced bone destruction. Because mature osteoclasts also express MIP-1δ receptors, MIP-1δ may have an effect on osteoclasts survival. In addition, evidence suggests that the MIP-1δ family member, MIP-1α, is capable of stimulating RANKL expression in bone mar row stromal cells in vitro (9). Should MIP-1δ elicit a similar effect, this may represent an indirect mechanism through which MIP-1δ could stimulate osteoclast differentiation. Lastly, although we have provided evidence that MIP-1δ may influence RBM-induced osteolysis through effects on bone-resorbing osteoclasts, MIP-1δ may also have effects on bone-forming osteoblasts. Although evidence suggests that bone loss observed in RBM patients is mediated largely by increased bone destruction (10), reduced bone formation may also play a role. Because osteoblasts express both MIP-1δ receptors CCR1 and CCR3,1

1

Our unpublished observations.

it will be important to examine the effect of MIP-1δ on osteoblast cell viability and differentiation as well.

Patients with RBM have a dismal prognosis with <10% surviving 5 years (11). RBM is highly resistant to standard therapies and, although surgical management can palliate the morbidity associated with osteolytic bone lesions, it cannot slow the progression of the disease. It is widely held that the liberation of growth factors stored in the bone via bone resorption plays a significant role in supporting the growth of metastatic bone tumors. Therefore, the successful treatment of RBM is likely to require a combination of therapies to directly reduce tumor viability and simultaneously inhibit bone destruction. Our preliminary findings demonstrating the increased expression of the chemokine MIP-1δ in RBM and its ability to promote the recruitment and differentiation of osteoclast precursors suggest that it may serve as a novel therapeutic target. In addition, MIP-1δ may also be useful as a prognostic indicator. Although the average expression of MIP-1δ was higher in RBM tissues relative to that in primary RCC tissues, individual expression levels in primary RCC tissues were variable. In two of five primary RCC tissues tested, MIP-1δ was expressed at similar levels to that observed in RBM tissues, suggesting the possibility that a clinical correlation exists between MIP-1δ expression in the primary tumor and the development of bone metastasis. Future studies incorporating additional patient samples with long-term follow-up are needed to examine this possibility. Taken together, these data warrant further investigation into the biological role of MIP-1δ in RBM establishment and osteolysis, and its clinical potential as a prognostic indicator and therapeutic target.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

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.

We thank Dr. G. David Roodman for technical assistance and insightful discussion.

1
Zekri J, Ahmed N, Coleman RE, Hancock BW. The skeletal metastatic complications of renal cell carcinoma.
Int J Oncol
2001
;
19
:
379
–82.
2
Bendre MS, Gaddy-Kurten D, Mon-Foote T, et al. Expression of interleukin 8 and not parathyroid hormone-related protein by human breast cancer cells correlates with bone metastasis in vivo.
Cancer Res
2002
;
62
:
5571
–9.
3
Bendre MS, Montague DC, Peery T, Akel NS, Gaddy D, Suva LJ. Interleukin-8 stimulation of osteoclastogenesis and bone resorption is a mechanism for the increased osteolysis of metastatic bone disease.
Bone
2003
;
33
:
28
–37.
4
Choi SJ, Cruz JC, Craig F, et al. Macrophage inflammatory protein 1-α is a potential osteoclast stimulatory factor in multiple myeloma.
Blood
2000
;
96
:
671
–5.
5
Han JH, Choi SJ, Kurihara N, Koide M, Oba Y, Roodman GD. Macrophage inflammatory protein-1α is an osteoclastogenic factor in myeloma that is independent of receptor activator of nuclear factor κB ligand.
Blood
2001
;
97
:
3349
–53.
6
Youn BS, Zhang SM, Lee EK, et al. Molecular cloning of leukotactin-1: a novel human β-chemokine, a chemoattractant for neutrophils, monocytes, and lymphocytes, and a potent agonist at CC chemokine receptors 1 and 3.
J Immunol
1997
;
159
:
5201
–5.
7
Weber KL, Pathak S, Multani AS, Price JE. Characterization of a renal cell carcinoma cell line derived from a human bone metastasis and establishment of an experimental nude mouse model.
J Urol
2002
;
168
:
774
–9.
8
Kusano K, Miyaura C, Inada M, et al. Regulation of matrix metalloproteinases (MMP-2, -3, -9, and -13) by interleukin-1 and interleukin-6 in mouse calvaria: association of MMP induction with bone resorption.
Endocrinology
1998
;
139
:
1338
–45.
9
Abe M, Hiura K, Wilde J, et al. Role for macrophage inflammatory protein (MIP)-1α and MIP-1β in the development of osteolytic lesions in multiple myeloma.
Blood
2002
;
100
:
2195
–202.
10
Aoki J, Yamamoto I, Hino M, et al. Osteoclast-mediated osteolysis in bone metastasis from renal cell carcinoma.
Cancer
1988
;
62
:
98
–104.
11
Russo P. Renal cell carcinoma: presentation, staging, and surgical treatment.
Semin Oncol
2000
;
27
:
160
–76.

Supplementary data