Nobuo Kugai

Bisphosphonates act on osteoblastic cells and inhibit osteoclast formation in mouse marrow cultures

We examined the mode of action of bisphosphonates on osteoclastic cell recruitment using mouse marrow cultures with or without osteoblastic cells. Tartrate-resistant acid phosphatase-positive multinucleated cells [TRAP(+)MNC] formed in cultures were determined to be osteoclastic cells. In marrow cultures, TRAP(+) MNC formation in the presence of 10(-8) mol/L 1,25(OH)2D3 was not affected by the addition of 10(-6) mol/L dihydrogen (cycloheptylamino)-methylenebisphosphonate monohydrate (YM175). However, it was inhibited in cocultures of marrow cells with osteoblastic cells. The inhibitory effect was evident throughout the entire culture period. YM175 dose dependently inhibited TRAP(+) MNC formation, and other bisphosphonates--pamidronate and alendronate--also inhibited TRAP(+) MNC formation in the coculture. Similar observations were also made in the coculture of spleen cells with osteoblastic cells. The conditioned media of osteoblastic cells treated with 10(-6) mol/L YM175 inhibited TRAP(+) MNC formation in marrow cultures. The presence of YM175 in methylcellulose cultures affected neither the colony formation of monocyte-macrophage lineage, nor TRAP(+) MNC formation in the succeeding cocultures of recovered cells with osteoblastic cells. These results indicate that YM175 and probably other bisphosphonates as well preferentially inhibit the later stage of osteoclastogenesis through its action on osteoblastic cells. Our findings suggest that part of the inhibitory action by osteoblastic cells in the presence of bisphosphonates is mediated through soluble factor(s).

Involvement of cyclo-oxygenase-2 in osteoclast formation and bone destruction in bone metastasis of mammary carcinoma cell lines.

We previously reported that mouse mammary carcinoma cell lines (MMT060562 and BALB/c‐MC) induced osteoclast formation through production of prostaglandin E2 (PGE2) in cocultures with mouse bone marrow cells, but the mechanism(s) of PG production remained unclear. In the present in vitro and in vivo studies, we tested the involvement of cyclo‐oxygenase‐2 (COX‐2), an inducible rate‐limiting enzyme in PG biosynthesis, in the stimulation of osteoclast formation by mouse mammary carcinoma cell lines. Addition of a selective COX‐2 inhibitor, JTE‐522, to cocultures of mammary carcinoma cell lines and bone marrow cells lowered PGE2 concentration in the culture media and inhibited osteoclast formation in a dose‐dependent manner. Northern blotting showed a very high level of COX‐2 messenger RNA (mRNA) expression in MMT060562. The mRNA expression was low in BALB/c‐MC, but it increased when BALB/c‐MC and bone marrow cells were cocultured. The results of immunocytochemistry for COX‐2 protein in respective cultures were compatible with the results of COX‐2 mRNA. In vivo, BALB/c‐MC injected into the heart of Balb/c mice metastasized to bone and formed osteolytic lesions in their hindlimbs. Histological examination revealed that tumor cells had metastasized to the bone marrow cavity and destroyed the bone trabeculae. Immunohistochemistry demonstrated that bone marrow stromal cells adjacent to tumor cells expressed COX‐2 protein. These findings suggest that COX‐2 plays an important role in the osteolysis of bone metastasis in vivo as well as in osteoclast formation in cocultures used as an in vitro model of metastatic bone disease.

Biphasic Effect of Prostaglandin E2 on Osteoclast Formation in Spleen Cell Cultures: Role of the EP2 Receptor†

We examined the effect of PGE2 on OC formation from spleen cells treated with M‐CSF and RANKL. PGE2 decreased OC number at 5–6 days of culture and increased OC number, size, and resorptive activity at 7–8 days. A selective EP2 receptor agonist mimicked these effects. Deletion of the EP2 receptor or depletion of T‐cells abrogated the increase in OC number.

Urinary type IV collagen as a marker for early diabetic nephropathy

We investigated the urinary secretion of type IV collagen in 115 subjects with non-insulin-dependent diabetes mellitus (NIDDM) without macroproteinuria, 34 normal healthy subjects and 19 subjects with chronic glomerulonephritis (CGN). We examined the relation between the urinary level of type IV collagen and various clinical parameters. The urinary level of type IV collagen was significantly elevated in NIDDM subjects compared with normal subjects (4.88 +/- 3.12 vs. 1.7 +/- 1.25 micrograms/gCr, P < 0.001). The urinary level of type IV collagen was increased even in NIDDM subjects with normoalbuminuria. The ratio of urinary type IV collagen was significantly lower in subjects with chronic glomerulonephritis (CGN) than those in NIDDM subjects (P < 0.001), although there was no significant difference in the urinary level of type IV collagen between NIDDM and CGN subjects. The ratio of urinary type IV collagen to albumin was under 10.0 x 10(-6) in all subjects with CGN. Our results suggest that measurement of the urinary level of type IV collagen is useful for detection of early diabetic nephropathy and for the differential diagnosis of diabetic nephropathy and chronic glomerulonephritis.

The Mouse Mammary Tumor Cell Line, MMT060562, Produces Prostaglandin E2 and Leukemia Inhibitory Factor and Supports Osteoclast Formation In Vitro Via a Stromal Cell–Dependent Pathway

Osteoclastic bone resorption increases at the site of bone metastasis, but little is known about how tumor cells induce osteoclast (OC) recruitment in the bone marrow microenvironment. To clarify this point, we examined the effects of various mouse tumor cells on OC recruitment using cocultures of tumor cells and mouse marrow cells. The mouse mammary tumor cell lines, MMT060562 (MMT), BALB/c‐MC, Jyg‐MC(A), or other nonmammary tumor cell lines, LLC and B16, were cocultured with mouse marrow cells, and OC recruitment from marrow cells was determined by counting the number of tartrate‐resistant acid phosphatase–positive multinucleated cells (TRAP(+) MNCs) formed. Of the tumor cells examined, MMT and BALB/c‐MC stimulated OC formation, but other tumor cells did not. OC formation with MMT was dependent on the number of MMTs inoculated, and only ten cells per well were sufficient to induce OC development. OCs appeared on day 4, and the number reached a maximum on days 5–8 and decreased thereafter. TRAP(+) MNCs induced by MMT satisfied the major criteria of OCs, such as the presence of calcitonin receptors and the ability to resorb calcified tissues. The majority of OCs were formed adjacent to the stromal cells, which were positive for alkaline phosphatase. When spleen cells were cocultured with MMT, no OCs were formed. In contrast, when osteoblastic cells were added to cocultures of spleen cells and MMT, many OCs were formed. The cultured media (CM) of MMT induced OC formation in mouse marrow cultures. Neither parathyroid hormone–like nor interleukin 1‐like activity was present in the CM. MMT constitutively produced prostaglandin E2 (PGE2) and OC formation in cocultures was completely inhibited by indomethacin. Fractionation of the CM of MMT by ultrafiltration indicated that the OC‐inducing activities were present not only in the fraction with molecular weight below 3 kDa but also in the fraction with molecular weight above 3 kDa. OC‐inducing activity with high molecular weight was eluted around 50 kDa by Bio‐Gel P‐60 column chromatography. The active fractions also possessed leukemia inhibitory factor (LIF) activity, and OC‐inducing activity of the peak fraction was inhibited in the presence of anti‐LIF neutralizing antibody. The results of this study indicated that MMTs release PGE2 and LIF, which in turn stimulate OC formation via a stromal cell–dependent pathway. These culture systems will help to clarify the mechanisms by which tumor cells induce OC formation in a bone marrow microenvironment.

The effect of deletion of cyclooxygenase-2, prostaglandin receptor EP2, or EP4 in bone marrow cells on osteoclasts induced by mouse mammary cancer cell lines1

The inducible prostaglandin (PG) synthesis enzyme, cyclooxygenase-2 (COX-2), is involved in osteoclast (OC) formation in cocultures of mouse mammary cancer cell lines (MMT060562 or BALB/c-MC) and bone marrow cells through production of PGE(2). There are four PGE(2) receptors but only the EP2 and EP4 receptors are reported to be important for OC formation. We have investigated the role of COX-2, EP2 receptor, and EP4 receptor in marrow cells for osteoclastogenesis in cocultures of cancer cells and bone marrow cells. We cocultured cancer cell lines with bone marrow cells from COX-2 knockout (-/-), EP2 -/- or EP4 -/- mice compared to wild-type mice. In addition, an EP4 receptor antagonist (EP4 RA) was added in some cocultures. Disruption of COX-2 gene in bone marrow cells had no effect on PGE(2) production and OC formation in cocultures with MMT060562, while it abrogated PGE(2) production and OC formation in cocultures with BALB/c-MC. Disruption of the EP2 gene in bone marrow cells had no effect on OC formation in the cocultures, while disruption of the EP4 gene in bone marrow cells abrogated OC formation in the cocultures. Furthermore, EP4 RA suppressed OC formation and prevented the increase in receptor activator of nuclear factor kappaB ligand (RANKL) mRNA levels in the cocultures. We conclude that COX-2 in cancer cells is responsible for PGE(2) and OC production in cocultures with MMT060562, while COX-2 in bone marrow cells, not cancer cells, is responsible for PGE(2) and OC production in cocultures with BALB/c-MC, and EP4 receptors are essential for OC formation in both cocultures.

A Third‐Generation Bisphosphonate, YM175, Inhibits Osteoclast Formation in Murine Cocultures by Inhibiting Proliferation of Precursor Cells via Supporting Cell‐Dependent Mechanisms

The theory that bisphosphonates inhibit osteoclast formation through their effects on osteoblastic cells remains controversial. To confirm the inhibitory effect of bisphosphonates on osteoclast formation and gain some insights into the underlying mechanisms, we examined the effect of disodium dihydrogen (cycloheptylamino)‐methylenebisphosphonate monohydrate (YM175) on osteoclast‐like multinucleated cell (OCL) formation in various mouse coculture systems. When different origins of osteoclast precursors (bone marrow, spleen, or nonspecific esterase‐positive cells) were cocultured with the same supporting cells (calvarial osteoblasts), YM175 inhibited OCL formation similarly in all cultures. When the same osteoclast precursors (spleen cells) were cocultured with supporting cells of different origin, the results were variable. YM175 inhibited OCL formation almost completely in cocultures with calvarial osteoblasts or osteoblastic cell line KS4, while it did not, or only slightly, inhibit OCL formation in cocultures with stromal cell lines, ST2 or MC3T3‐G2/PA6. Temporal addition of YM175 in cocultures of spleen cells with osteoblastic cells revealed that YM175 was effective when it was present at an early phase of the culture period. Consistent with this observation, YM175 in the presence of osteoblastic cells inhibited proliferation of preosteoclastic cells, but did not inhibit the fusion of mononuclear prefusion osteoclasts. In conclusion, the inhibitory effect of YM175 on OCL formation was confirmed in various murine coculture systems, but the effect was dependent on the types of bone‐derived cells supporting osteoclastogenesis. The findings suggest that YM175 inhibits osteoclastogenesis by inhibiting the proliferation of osteoclast precursors through its action on supporting cells of osteoblast lineage rather than acting directly on osteoclast precursors.

Mouse mammary carcinoma cell line (BALB/c-MC) stimulates osteoclast formation from mouse bone marrow cells through cell-to-cell contact

We recently reported that numerous osteoclasts (OC) were formed in cocultures of some mouse cancer cell lines and bone marrow cells. In this study, we examined mechanisms by which one of the cell lines, BALB/c-MC, induces OC. BALB/c-MC dose dependently stimulated OC formation in cocultures. In cocultures where direct cell-to-cell contact between BALB/c-MC and bone marrow cells was inhibited by membrane filters, OC formation was not stimulated. The stimulation of OC formation in the coculture was completely abolished by adding 10(-7)-10(-6) mol/L indomethacin. The concentration of prostaglandin E2 (PGE2) in the culture media of cocultures with cell-to-cell contact was higher than that of cocultures without cell-to-cell contact or marrow cultures alone, and it reached levels sufficient to induce OC (11.9 +/- 5.3 ng/mL [about 3.4 x 10(-8) mol/L]). When BALB/c-MC or bone marrow cells were fixed with formalin and then cocultured with bone marrow cells or BALB/c-MC, respectively, the concentration of PGE2 in the culture media of cocultures of fixed BALB/c-MC and bone marrow cells increased, whereas that of cocultures of BALB/c-MC and fixed bone marrow cells did not increase. These results indicate that BALB/c-MC stimulate OC formation through direct cell-to-cell contact with bone marrow cells, and PGE2 released from bone marrow cells through direct cell-to-cell contact are involved in OC formation by the cell line.

Enhancement of cyclic adenosine monophosphate content in bone cells by the factor extracted from a pancreatic cancer associated with hypercalcemia

A woman with exocrine pancreatic cancer presented a syndrome of humoral hypercalcemia of malignancy (HHM). Either urea extract or acid/ethanol extract of the tumor showed a dose-dependent activity to elevate cyclic adenosine monophosphate (AMP) level in rat bone cells in primary culture. When each population obtained by the sequential digestion of rat fetal calvaria was cultured individually and cyclic AMP responses to parathyroid hormone (PTH), calcitonin, and tumor extract were examined, tumor extract-sensitive cells showed a similar distribution to PTH-sensitive cells. Tumor extract and PTH, but not calcitonin, increased cyclic AMP in osteogenic cell line MC 3T3-E1. PTH receptor-mediated increase of cyclic AMP was indicated by an antagonistic action of PTH analogue, (3-34) hPTH, on increase of cyclic AMP in MC 3T3-E1 elicited by tumor extract. Human breast cancer derived cell line MCF-7 had calcitonin-sensitive adenylate cyclase, but neither PTH nor tumor extract increased cyclic AMP in the cells. On Bio-Gel P-60 column, the activity to stimulate bone cell cyclic AMP was eluted as a single peak at the molecular size between 6.5 K and 12.4 K. It was concluded that pancreatic cancer, although rather exceptional as a cause of HHM, produced a factor very similar to that reported in representative HHM tumors of human and animal models.

Chinese Hamster Ovary Cells Expressing α4β1 Integrin Stimulate Osteoclast Formation In Vitro

It is reported that Chinese hamster ovary cells transfected with human α4 cDNA (α4CHOs) and expressing functional α4β1 integrin developed bone metasasis in nude mice. To clarify the role of α4β1 integrin in bone metastasis, in terms of tumor‐mediated bone destruction, we examined whether α4CHOs stimulate osteoclast formation in cocultures with mouse bone marrow cells. The number of osteoclast‐like cells identified as tartrate‐resistant acid phosphatase positive multinucleated cells (TRAP(+) MNCs) formed from bone marrow cells increased with the increasing number of α4CHOs cocultured. The effects of 1,25‐dihydroxyvitamin D3(1,25(OH)2D3) and prostaglandin E2 (PGE2) on TRAP(+) MNC formation were enhanced in cocultures with α4CHOs. TRAP(+) MNCs induced by α4CHOs possessed calcitonin receptors and resorbed calcified tissues. In cocultures, α4CHOs and bone marrow stromal cells were in contact with each other and bone marrow stromal cells expressed vascular cell adhesion molecule‐1 (VCAM‐1), which is one of the ligands for α4β1 integrin. TRAP(+) MNC formation was not stimulated in cocultures where direct contact between α4CHOs and bone marrow cells was inhibited by membrane filters. α4CHOs do not support TRAP(+) MNC formation in cocultures with spleen cells but do support TRAP(+) mononuclear cell and MNC formation from spleen cells in the presence of osteoblastic cells. Cultured media from α4CHOs, bone marrow cells, and cocultures of α4CHOs and bone marrow cells did not stimulate TRAP(+) MNC formation or enhance the effects of 1,25(OH)2D3 and PGE2 in bone marrow cultures. The concentrations of PGE2 and interleukin‐6 (IL‐6) in cultured media were not different between the cultures of bone marrow cells and the cocultures of bone marrow cells and α4CHOs. Anti‐human α4 and anti‐mouse VCAM‐1 antibodies inhibited TRAP(+) MNC formation induced by α4CHOs. These results indicate that α4CHOs stimulated TRAP(+) MNC formation through direct cell‐to‐cell interaction between α4β1 and VCAM‐1. It is suggested that in addition to various soluble factors regulating osteoclast formation, cell‐to‐cell interaction between tumor cells and bone marrow cells is important for inducing osteoclasts at the site of bone metastasis and leading to bone destruction.