In normal B-cell ontogeny, interleukin-6 (IL-6) is an important growth factor in the terminal differentiation of B cells into immunoglobulin-secreting plasma cells. Normal B cells produce antibody in response to IL-6 but do not proliferate. In contrast, IL-6 has been shown to be a central growth factor for myeloma cells.[5] The fact that myeloma cells proliferate in response to IL-6 is a major difference that distinguishes malignant from normal plasma cells and is of critical importance in the pathogenesis of the disease.

Early work by Potter[6,7] demonstrated that paraffin oil or pristane induced plasmacytomas when injected into BALB/c mice. The generation of the plasmacytomas was dependent on factors produced by the inflammatory cells. These cells were subsequently shown to produce IL-6, a potent growth factor for plasmacytomas.[6] More

recently, transgenic mice (C57BL/6) carrying the human IL-6 gene fused to a human immunoglobulin heavy chain enhancer developed a massive lethal plasmacytosis.[8,9]

Kishimoto and colleagues[10] have demonstrated that IL-6 is an autocrine growth factor for human myeloma cells. They have shown that myeloma cells freshly isolated from patients produce IL-6 and express its receptor. Exogenous IL-6 augments the in vitro growth of myeloma cells, and anti-IL-6 antibody inhibits their growth.[10] Schwab et al[11] have demonstrated that a myeloma cell line U266 expresses mRNA for both IL-6 and IL-6R. The proliferation of this cell line can be inhibited using anti-IL-6 antibody or

antisense IL-6 oligonucleotides further supporting the critical role of IL-6 in the growth of these cells.[11] Significantly elevated serum IL-6 levels have been detected in 3% of MGUS/SMM patients, in 35% of overt myeloma patients, and in 100% of a plasma cell leukemia group.[12] Using an antibromodeoxyuridine monoclonal antibody to specifically count myeloma cells in S-phase (ie, the labeling index), the IL-6 responsiveness of myeloma cells in vitro correlates with their labeling index in vivo and hence to the severity of the disease.[13]

Most importantly, an antibody to IL-6 administered in vivo has been shown to dramatically decrease the labeling index of the tumor cells in five patients with aggressive MM.[14,15]

Since IL-6 is a central growth factor for myeloma cells, sIL-6R may modulate IL-6 activity. Soluble receptors have been shown to be potent immunomodulators of their respective ligands. We have previously reported a novel IL-6R mRNA from myeloma cells that exhibits a 94-nt deletion of the entire transmembrane domain from codons 356 (G-TG) to 387 (AG-G).[16] The transmembrane domain deletion results in a shift in the translational reading frame with the insertion of 10 new amino acids followed by a stop codon. Sequence analysis shows the ligand-binding domain of the sIL-6R to be identical to that of the membrane-bound IL-6R up to the transmembrane domain deletion.

The sIL-6R cDNA was expressed and supernates were collected from mock or sIL-6R transfected PA-1 cells after 48 hours and assayed for their ability to stimulate or suppress the growth of an IL-6-dependent cell line, ANBL-6. Soluble IL-6R alone had no effect on

the growth of the ANBL-6 cells. However, the growth of ANBL-6 cells by sIL-6R was potentiated in the presence of IL-6 and could be blocked by anti-IL-6 antibody.[17] The above results suggest that, in the presence of IL-6, sIL-6R associates with gp130 leading to signal transduction and cell growth.

Among 30 healthy individuals, 32 patients with MGUS, and 74 with myeloma, sIL-6R levels were increased similarly in MGUS and MM -- 51% and 44%, respectively.[18] An elevated sIL-6R level correlated with a poor survival and was independent of the plasma-cell labeling index and [Image]2M. Soluble IL-6R plays an important role in the pathogenesis of MM by potentiating IL-6 activity.

Although it is clear that IL-6 expression plays a fundamental role in the growth of MM cells, the source of IL-6 expression is controversial. Klein et al[19] reported that the high levels of IL-6 found in the bone marrow of patients with progressive MM is confined to the adherent cells of the bone marrow environment and that IL-6 is not expressed by myeloma cells. In addition, bone marrow monocytic and myeloid cells, but not myeloma cells, have been reported to express IL-6 mRNA.[20] Together, these results suggest a paracrine rather than an autocrine mechanism of myeloma cell growth by IL-6. As discussed above, Kishimoto and colleagues have demonstrated that IL-6 is an autocrine growth factor for human myeloma cells.

Hata et al[21] detected IL-6 mRNA by RT/PCR in purified plasma cells from myeloma patients as well, and they also demonstrated that these CD38+ myeloma cells expressed intermediate levels of CD45. Our data demonstrate that monoclonal plasma cells from the majority of myeloma patients with active disease manifested by a high labeling index can express IL-6 mRNA in an autocrine fashion.[22] Furthermore, we have reported three new myeloma cell lines isolated from patients with very high labeling indices that produce IL-6 in an autocrine fashion, two of which are CD38+/CD45+.[23] The apparent discrepancy between our data and those of Klein et al[19] is likely due to the differences in sensitivities of the techniques used to detect IL-6 expression; we utilized polymerase chain reaction, which is more sensitive than Northern analysis. However, it is likely that both autocrine and paracrine sources of IL-6 production play a role in the pathogenesis of myeloma.

Production of IL-1[Image] by myeloma cells may be responsible for the paracrine generation of IL-6 by marrow stromal cells. It has been shown that IL-1[Image] can induce expression of the genes for IL-6, colony-stimulating factors, and adhesion proteins. In vitro fibroblasts, macrophages, T lymphocytes, and other marrow stromal cells are all capable of responding to IL-1[Image] by expressing one or more IL-1[Image]

inducible genes.[24] Carter et al[25] have found that human myeloma cells are able to induce IL-6 production in marrow stromal cells. The stimulatory activity of the myeloma cells appears to be mediated through endogenously released IL-1[Image] and antibodies to IL-1[Image] completely abrogate the IL-6 production.

Normal plasma cells do not produce IL-1 ; however, abnormal IL-1 production by myeloma cells has been detected at both the mRNA and protein levels by several different investigators. Using fresh myeloma cells, Lichtenstein and colleagues[26] detected IL-1[Image] at the rotein level, and Klein et al[27] found strong IL-1[Image] gene expression by in situ ybridization. Cozzolino et al[28] have shown that culture supernatants of plasma cells, solated by a rosetting procedure, from 12 of 12 patients with MM contained high amounts of IL-1[Image]. In contrast, plasma cells from 9 of 9 patients with MGUS showed undetectable levels of IL-1[Image]. Using flow cytometric sorting to enrich for plasma cells and RT/PCR for cytokine expression, we have found that IL-1[Image] mRNA is expressed by plasma cells from virtually all MM patients but is not detectable in the plasma cells of most MGUS patients.[22] Future studies will determine whether the detection of IL-1[Image] xpression will differentiate between patients with MGUS or MM.

myeloma? The development of osteolytic lesions is an important clinical finding that clearly distinguishes MGUS from myeloma.[1] ince IL-1[Image] has potent osteoclast-activating factor (OAF) activity, t may be responsible for the presence of bone lesions. Initially, two different groups had shown that the bone resorbing activity in supernatants of myeloma cell cultures was likely due to IL-1[Image] and not to IL-1 [Image], tumor necrosis factor, or lymphotoxin.[27,29] More recently, Torcia and colleagues[30] have shown a critical role for IL-1[Image] in the pathogenesis of bone disease. Using the fetal rat long-bone tissue culture assay, they demonstrated that the OAF activity of culture supernatants from nfractionated bone marrow cells from myeloma patients correlated with the IL-1[Image] content (r = 0.949). Furthermore, the OAF activity could be completely abolished by IL-1ra, sIL-1R type I or II, or neutralizing nti-IL-1[Image] antibodies but not anti-IL-6 antibodies. These results demonstrate that the OAF activity of myeloma cells from patients is predominantly, if not solely, related to IL-1[Image].[30]

Hawley and colleagues[31] developed a mouse model of myeloma that demonstrates the importance of IL-1 expression in inducing pathology that mimics human disease. They introduced an IL-1 cDNA into an IL-6 dependent murine B-cell line by retroviral-mediated gene transfer. After injection of these IL-1 producing B-cells into syngeneic mice, these cells were shown to "home" to the bone marrow and produce metastatic bone lesions. By comparison, intravenous injection of autonomously growing B-cell lines generated in vitro by retroviral insertion of an IL-6 cDNA rarely resulted in bone marrow or bone metastases.[31] Subsequent work has shown that aberrant expression of IL-1 can alter adhesion molecules such as ICAM and CD44 on the surface of mouse plasmacytoma cells.[32] A similar mechanism may occur in human myeloma in which aberrant expression of IL-1[Image] induces increased expression of adhesion molecules such as VLA-4, CD44, CD54, CD56, and other surface molecules on the monoclonal plasma cells.[33-37]