We read the publication by Farrugia et al. (1) with great interest. A novel system of cytokines was identified as key mediators in osteoclastogenesis, including the receptor activator of nuclear factor κB ligand (RANKL), receptor activator of nuclear factor κB, and osteoprotegerin (2). RANKL binds to its specific receptor, receptor activator of nuclear factor κB, which is located on osteoclastic precursors and induces osteoclastogenesis. Osteoprotegerin acts as a decoy receptor for RANKL and neutralizes its biological effects. Farrugia et al. (1) reported that RANKL is expressed by human myeloma cells, and this expression correlates with bone destruction. This publication is important because there has been a controversy whether myeloma cells express RANKL (3). Whereas several authors reported RANKL expression by myeloma cells (4, 5, 6), no expression could be detected by others using immunohistochemistry (7, 8). We have used immunocytochemistry rather than immunohistochemistry to show the RANKL expression of human myeloma cells in bone marrow aspirates (9). The reason was that the immunohistochemical evaluation of formalin-fixed, decalcified, and paraffin-embedded tissue sections may be associated with limitations because it is known that decalcification of formalin-fixed tissue samples reduces its antigenicity. Bypassing the need for formalin fixation and decalcification, we could detect a strong cytoplasmatic expression of RANKL in bone marrow plasma cells in all of the investigated multiple myeloma patients with lytic bone disease (9). These results are in line with the observation by Farrugia et al. (1) that the epitope identified by the monoclonal RANKL antibody was sensitive to paraformaldehyde-fixation.

Farrugia et al. (1) found a significantly higher RANKL expression on plasma cells from patients with multiple myeloma and lytic bone disease in comparison to patients without osteolysis. Plasma cells were defined as CD38+++ cells using two-color flow cytometry. We had used three-color flow cytometry in bone marrow aspirates from 50 patients with multiple myeloma with CD38 PC5, CD138 FITC, and a monoclonal antihuman RANKL antibody conjugated to PE (10). Plasma cells were identified as cells strongly positive for CD38 and coexpressing CD138. The bone marrow plasma cells from controls showed no or only a weak surface expression of RANKL, and the median mean fluorescence index was 6. The expression of RANKL on bone marrow plasma cells from myeloma patients was significantly correlated with the presence (median mean fluorescence index = 60) or absence (median mean fluorescence index = 16) of bone lesions (P < 0.0005; Ref. 10). Additionally we isolated primary bone marrow plasma cells using another method than Farrugia et al. (1), namely immunomagnetic cell sorting using an antihuman CD138 antibody conjugated to magnetic beads (CD138 MicroBeads; Miltenyi Biotec, Bergisch Gladbach, Germany). RANKL mRNA could be detected in freshly isolated bone marrow plasma cells from myeloma patients with osteolytic myeloma bone disease. These data show that human bone marrow myeloma cells directly produce RANKL on mRNA and protein level, and the level of RANKL protein expression on myeloma cells correlates with lytic bone disease in patients with multiple myeloma.

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.

Reply

We appreciate the careful and critical reading of our article by Drs. Giuliani and Sezer and the opportunity to respond to their comments. As highlighted in the letter from Dr. Sezer, the topic of our article (1) has been the subject of much controversy in the field of myeloma biology (2,3,4,5,6) and centers on whether myeloma plasma cells express the potent pro-osteoclastogenic molecule receptor activator of nuclear factor κB ligand (RANKL) or secrete factor(s) that cause an increase in RANKL expression by surrounding stromal cells/T cells in the bone marrow (2,3,4,5,6). Whereas our article in no way precludes the latter scenario, it does show by a variety of methodologies that myeloma plasma cells express RANKL mRNA and functional protein. Moreover, it shows that myeloma plasma cells also express the soluble isoform of RANKL, shown previously to be expressed by activated T cells in inflammatory conditions such as rheumatoid arthritis (7, 8). Consistent with work of Heider et al. (9), our studies show that RANKL protein expression by plasma cells is associated significantly with the presence of osteolytic lesions in patients, highlighting the potential prognostic and therapeutic value of our findings.

We acknowledge the concerns raised by Giuliani et al., but feel that a number of their comments are inconsistent with recent published findings (6, 9). Firstly, Giuliani et al. quoted data from their laboratory (2, 3) and others (4, 10) who show that patient-derived CD138-positive plasma cells and human myeloma cell lines do not express RANKL mRNA and protein. In contrast, our studies unequivocally show that the human myeloma cell lines RPMI-8226, U266, WL-2, and the EBV-positive ARH-771 as well as patient-derived myeloma cells all express RANKL mRNA and protein (1). Whereas initially reporting that plasma cells in bone marrow trephines lacked RANKL expression (4), Yaccoby et al. (11) have reported recently that CD138-isolated human plasma cells express membrane-associated RANKL. Furthermore, studies by Croucher et al. (12) show that the widely used murine myeloma cell line model, 5TMM, expresses surface RANKL. Whereas we are unable to explain the microarray studies published by Zhan et al. (13), we postulate that this may relate to the relatively low level of RANKL transcript within these cells and the relative instability of this transcript. In considering the biological significance of this low level of RANKL expression, it must be remembered that it is the relative expression of RANKL to its inhibitor osteoprotegerin, which is critical in determining the osteoclastogenic effect. To this end, our studies also clearly show that CD38+++ plasma cells do not express detectable levels of osteoprotegerin mRNA (1). Therefore, in situations of significant plasma cell burden within the bone marrow microenvironment, this would result in a high ratio of RANKL to osteoprotegerin expression and a resultant pro-osteoclastogenic environment.

Secondly, Giuliani et al. question our use of CD38 expression to isolate myeloma plasma cells instead of the “widely accepted specific marker of myeloma cells,” CD138. Our experience, as with a number of other groups, shows that CD138 is variably expressed and sometimes lacking in patients with significant plasma cell burden (14, 15). On this basis, we chose to use high expression of CD38 to isolate bone marrow-derived plasma cells. We strongly reject the suggestion that our results are due to “other contaminating cells that are known to express RANKL such as activated T lymphocytes,” because the purity of all of our sorted populations was routinely assessed and found to be >99%. Whereas we recognize that activated T lymphocytes are capable of expressing the CD38 antigen (16), we as others have reported (14), find no evidence of contaminating CD3+ T lymphocytes in the CD38+++ population. Although we have not examined RANKL mRNA expression in sorted CD138 plasma cells, studies by Sezer et al. suggest that magnetic activated cell sorting purified CD138+ cells do indeed express RANKL mRNA (9). Our two-color flow cytometric analysis on bone marrow derived from several myeloma patients show that CD138+ cells react with two anti-RANKL antibodies, MAB626 (R& D Systems) and polyclonal anti-RANKL, Sc-9073 (Santa Cruz Biotechnology). Once again, these findings are consistent with findings from the laboratory of Dr. Sezer as outlined in his letter. Furthermore, using flow cytometry, recent studies from our laboratory show that sf21-derived huRANKL (the immunogen used to raise MAB626) can block the binding of MAB626 to the human myeloma cell lines 8226, confirming the specificity of this monoclonal antibody (Fig. 1).

With regard to the ability of myeloma plasma cells to support osteoclast differentiation and activation, it seems that Giuliani et al. may have misinterpreted the data presented. The CD38+ cells, which also express RANKL, are less capable of supporting TRAP+ osteoclast development, when compared with the CD38+++ plasma cell population. This, in our opinion, reflects the fact that the CD38+ population also expresses mRNA encoding osteoprotegerin. Whereas our laboratory routinely performs osteoclast-forming assays with peripheral blood derived magnetic activated cell sorting separated CD14+ monocytes, we find little difference in the phenotype and functional capacity of magnetic activated cell sorting selected CD14+ osteoclast precursors and those obtained by adherence as described in the article (1). Moreover, unlike what is suggested by Giuliani et al., we find no evidence of contaminating CD3+ T lymphocytes in our osteoclast precursor population. While we recognize that when compared with stromal cells or, as was the case in our study, compared with osteoclasts generated by continuous exposure to recombinant RANKL, the “osteoclastogenic effect” observed “is little” when osteoclast precursors are cultured with the CD38+++ cells, it nevertheless demonstrates that myeloma plasma cells are capable of supporting osteoclast formation and activation in a RANKL-dependent manner. Indeed, we believe that the more modest osteoclast formation observed when CD38+++ cells were used as the stromal support may relate in part to the poor in vitro survival capacity of the primary plasma cells used (1).

We feel that our results reflect the careful nature in which our study was performed and as highlighted in the letter by Sezer, represents one of a number of recent studies that will recognize the importance and the role of plasma cell-derived RANKL in myeloma biology.

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.

Requests for reprints: Andrew C. W. Zannettino, Myeloma and Mesenchymal Research Group, Matthew Roberts Foundation Laboratory, Division of Haematology, Institute of Medical and Veterinary Science and The Hanson Institute, Frome Road, Adelaide, South Australia 5000, Australia. Phone: 61-8-8222-3455; Fax: 61-8-8222-3139; E-mail: andrew.zannettino@imvs.sa.gov.au

1

A. N. Farrugia and A. C. W. Zannettino, unpublished observations.

Fig. 1.

The expression of membrane-associated receptor activator of nuclear factor κB ligand (RANKL) protein by the HMCL 8226 was measured using indirect immunofluorescence and flow cytometry. Two hundred thousand 8226 cells were stained with a 100 μl solution of MAB626 at a concentration of 10 μg/ml (anti-RANKL; R&D Systems). Specific binding was revealed by incubation with saturating concentration of goat-antimouse IgG-FITC. The fluorescence histograms in A show the expression of RANKL protein (black histogram) compared with isotype-matched, nonbinding control used under identical conditions (1A6.11, IgG2b; gray histogram). In B, 100 μl solutions of MAB626 and 1D4.5, both at a concentration of 10 μg/ml, were incubated with 300 ng/ml SF-21-derived RANKL protein (R&D Systems) for 1 h at 37°C for 60 min before staining the HMCL 8226 as in A. The reactivity of MAB626 is almost completely inhibited by prior incubation with recombinant RANKL as evidenced by the reduction in transmembrane RANKL expression (A; MnX = 33) compared with untreated cells (B; MnX = 11).

Fig. 1.

The expression of membrane-associated receptor activator of nuclear factor κB ligand (RANKL) protein by the HMCL 8226 was measured using indirect immunofluorescence and flow cytometry. Two hundred thousand 8226 cells were stained with a 100 μl solution of MAB626 at a concentration of 10 μg/ml (anti-RANKL; R&D Systems). Specific binding was revealed by incubation with saturating concentration of goat-antimouse IgG-FITC. The fluorescence histograms in A show the expression of RANKL protein (black histogram) compared with isotype-matched, nonbinding control used under identical conditions (1A6.11, IgG2b; gray histogram). In B, 100 μl solutions of MAB626 and 1D4.5, both at a concentration of 10 μg/ml, were incubated with 300 ng/ml SF-21-derived RANKL protein (R&D Systems) for 1 h at 37°C for 60 min before staining the HMCL 8226 as in A. The reactivity of MAB626 is almost completely inhibited by prior incubation with recombinant RANKL as evidenced by the reduction in transmembrane RANKL expression (A; MnX = 33) compared with untreated cells (B; MnX = 11).

Close modal

References

21
Farrugia A. N., Atkins G. J., To L. B., Pan B., Horvath N., Kostakis P., Findlay D. M., Bardy P., Zannettino A. C. Receptor activator of nuclear factor-κB ligand expression by human myeloma cells mediates osteoclast formation in vitro and correlates with bone destruction in vivo.
Cancer Res.
,
63
:
5438
-5445,  
2003
.
22
Giuliani N., Colla S., Sala R., Moroni M., Lazzaretti M., La Monica S., Bonomini S., Hojden M., Sammarelli G., Barille S., Bataille R., Rizzoli V. Human myeloma cells stimulate the receptor activator of nuclear factor-κ B ligand (RANKL) in T lymphocytes: a potential role in multiple myeloma bone disease.
Blood
,
100
:
4615
-4621,  
2002
.
23
Giuliani N., Bataille R., Mancini C., Lazzaretti M., Barille S. Myeloma cells induce imbalance in the osteoprotegerin/osteoprotegerin ligand system in the human bone marrow environment.
Blood
,
98
:
3527
-3533,  
2001
.
24
Pearse R. N., Sordillo E. M., Yaccoby S., Wong B. R., Liau D. F., Colman N., Michaeli J., Epstein J., Choi Y. Multiple myeloma disrupts the TRANCE/ osteoprotegerin cytokine axis to trigger bone destruction and promote tumor progression.
Proc. Natl. Acad. Sci. USA
,
98
:
11581
-11586,  
2001
.
25
Sezer O., Heider U., Jakob C., Eucker J., Possinger K. Human bone marrow myeloma cells express RANKL.
J. Clin. Oncol.
,
20
:
353
-354,  
2002
.
26
Sezer O., Heider U., Jakob C., Zavrski I., Eucker J., Possinger K., Sers C., Krenn V. Immunocytochemistry reveals RANKL expression of myeloma cells.
Blood
,
99
:
4646
-4647, author reply 4647 
2002
.
27
Ikeda T., Kasai M., Utsuyama M., Hirokawa K. Determination of three isoforms of the receptor activator of nuclear factor-κB ligand and their differential expression in bone and thymus.
Endocrinology
,
142
:
1419
-1426,  
2001
.
28
Romas E., Gillespie M. T., Martin T. J. Involvement of receptor activator of NF[κ]B ligand and tumor necrosis factor-[α] in bone destruction in rheumatoid arthritis.
Bone
,
30
:
340
-346,  
2002
.
29
Heider U., Langelotz C., Jakob C., Zavrski I., Fleissner C., Eucker J., Possinger K., Hofbauer L. C., Sezer O. Expression of receptor activator of nuclear factor κB ligand on bone marrow plasma cells correlates with osteolytic bone disease in patients with multiple myeloma.
Clin. Cancer Res.
,
9
:
1436
-1440,  
2003
.
210
Roux S., Meignin V., Quillard J., Meduri G., Guiochon-Mantel A., Fermand J. P., Milgrom E., Mariette X. RANK (receptor activator of nuclear factor-κB) and RANKL expression in multiple myeloma.
Br. J. Haematol.
,
117
:
86
-92,  
2002
.
211
Yaccoby S., Shaughnessy J., Jr., Zhan F., Yi Q., Barlogie B., Epstein J. Role of myeloma-induced osteoclastogenesis in the disease.
Haematol. J.
,
4
:
S153
2003
.
212
Croucher P. I., Shipman C. M., Lippitt J., Perry M., Asosingh K., Hijzen A., Brabbs A. C., van Beek E. J., Holen I., Skerry T. M., Dunstan C. R., Russell G. R., Van Camp B., Vanderkerken K. Osteoprotegerin inhibits the development of osteolytic bone disease in multiple myeloma.
Blood
,
98
:
3534
-3540,  
2001
.
213
Zhan F., Tian E., Bumm K., Smith R., Barlogie B., Shaughnessy J., Jr. Gene expression profiling of human plasma cell differentiation and classification of multiple myeloma based on similarities to distinct stages of late-stage B-cell development.
Blood
,
101
:
1128
-1140,  
2003
.
214
Joshua D., Pope B., Brown R., Brown L., Murray A., Gibson J. Phenotyping primitive plasma cells.
Br. J. Haematol
,
117
:
252
-253,  
2002
.
215
Pope B., Brown R. D., Gibson J., Yuen E., Joshua D. B7–2-positive myeloma: incidence, clinical characteristics, prognostic significance, and implications for tumor immunotherapy.
Blood
,
96
:
1274
-1279,  
2000
.
216
Lima M., Teixeira Mdos A., Queiros M. L., Santos A. H., Goncalves C., Correia J., Farinha F. F., Mendonca F., Soares J. M., Almeida J. J., Orfao A., Justica B. Immunophenotype and TCR-Vβ repertoire of peripheral blood T-cells in acute infectious mononucleosis.
Blood Cells Mol. Dis.
,
30
:
1
-12,  
2003
.
1
Farrugia A. N., Atkins G. J., To L. B., Pan B., Horvath N., Kostakis P., Findlay D. M., Bardy P., Zannettino A. C. Receptor activator of nuclear factor-κB ligand expression by human myeloma cells mediates osteoclast formation in vitro and correlates with bone destruction in vivo.
Cancer Res.
,
63
:
5438
-5445,  
2003
.
2
Teitelbaum S. L. Bone resorption by osteoclasts.
Science (Wash. DC)
,
289
:
1504
-1508,  
2000
.
3
Sezer O., Heider U., Zavrski I., Kühne C. A., Hofbauer L. C. RANK ligand and osteoprotegerin in myeloma bone disease.
Blood
,
101
:
2094
-2098,  
2003
.
4
Altamirano C. V., Ma H. J., Parker K. M., Glowalla A., Mikail A. E., Sjak-Shie N., Vescio R., Berenson J. R. RANKL is expressed in multiple myeloma cell lines.
Blood
,
96
:
365a
2000
.
5
Croucher P. I., Shipman C. M., Lippitt J., Perry M., Asosingh K Hijzen A, Brabbs AC, van Beek EJ, Holen I., Skerry T. M., Dunstan C. R., Russell G. R., Van Camp B., Vanderkerken K. Osteoprotegerin inhibits the development of osteolytic bone disease in multiple myeloma.
Blood
,
98
:
3534
-3540,  
2001
.
6
Yaccoby S., Shaughnessy J., Zhan F., Yi Q., Barlogie B., Epstein J. Role of myeloma induced osteoclastogenesis in the disease. 9th International Multiple Myeloma Workshop.
Hematol. J.
,
4(Suppl. 1)
:
S153
2003
.
7
Pearse R. N., Sordillo E. M., Yaccoby S., Wong B. R., Liau D. F., Colman N., Michaeli J., Epstein J., Choi Y. Multiple myeloma disrupts the TRANCE/ osteoprotegerin cytokine axis to trigger bone destruction and promote tumor progression.
Proc. Natl. Acad. Sci. USA
,
98
:
11581
-11586,  
2001
.
8
Giuliani N., Bataille R., Mancini C., Lazzaretti M., Barille S. Myeloma cells induce imbalance in the osteoprotegerin/osteoprotegerin ligand system in the human bone marrow environment.
Blood
,
98
:
3527
-3533,  
2001
.
9
Sezer O., Heider U., Jakob C., Zavrski I., Eucker J., Possinger K., Sers C., Krenn V. Immunocytochemistry reveals RANKL expression of myeloma cells.
Blood
,
99
:
4646
-4647,  
2002
.
10
Heider U., Langelotz C., Jakob C., Zavrski I., Fleissner C., Eucker J., Possinger K., Hofbauer L. C., Sezer O. Expression of receptor activator of nuclear factor κB ligand on bone marrow plasma cells correlates with osteolytic bone disease in patients with multiple myeloma.
Clin. Cancer. Res.
,
9
:
1436
-1440,  
2003
.