The cause of multiple myeloma (MM) remains unexplained. However, important advances in the understanding of the genetic defects found in myeloma cells have been made. Syndecan-1 is found on the cell surface of most cells in the myeloma clone and binds growth factors such as fibroblast growth factor (FGF)-2. These findings provide important clues as to how the plasma cell becomes malignant and how it causes disease in patients.
When B lymphocytes are made by the bone marrow, they are stimulated by antigen in the lymph nodes into two kinds of plasma cellsÑthe pregerminal center plasma cell (pre-GCPC) and the post-germinal center plasma cell (post-GCPC). Pre-GCPCs are usually short lived. In contrast, post-GCPCs continue to make changes in their genetic structure so that they can bind even better to the antigen for which they are specific. These post-GCPCs become the long-lived PCs and typically migrate back to the bone marrow, where they live for months to years and manufacture their immunoglobulin (Ig). MM and monoclonal gammopathy of undetermined significance (MGUS) are tumors of long-lived post-GCPCs.
It is not clear what causes a long-lived PC to begin to proliferate and turn into active MM. It is unusual for patientsÕ MGUS to turn into MM (approximately 1% per year). One factor that is likely to be involved is instability of the genes in the PCs, referred to as "genetic instability." This can occur when B cells undergo three B-cell-specific DNA modification processes that generate double-strand DNA breaks Ð V (variable) D (diversity) J (joining) recombination, IgH switch recombination, and somatic hypermutation. Errors in any of these can produce a translocation.
Translocations of parts of one chromosome to another chromosome are common in MM. Usually, one partner of the exchange is a region on chromosome 14q32 where the genetic code for the heavy-chain portion of the antibody molecule is located. When a B cell is developing, it first makes IgM; later, when it matures into a PC, it switches its production to IgA or IgG. The region of 14q32 where these translocations occur is termed the "switch region."
Early in the development of a B cell, sequential, regulated DNA rearrangements are mediated by joining of segments of the VDJ genes. This process, referred to as "VDJ recombination," serves to generate IgH and IgL receptors. Later in development an antigen stimulates B cells to undergo IgH switch recombination, primarily in GC B cells, and also somatic hypermutation, which occurs exclusively in GC B cells. Mistakes in any of these three B-cellÐspecific DNA modifications can result in parts of chromosomes being moved to other chromosomes. In addition, somatic hypermutation can produce minor changes in small portions of the DNA, remove a section of DNA, or insert additional segments of DNA into a strand. These events can lead to abnormal cell activity and, hence, malignancy.
Dr. Michael KuehlÕs research at the National Cancer Institute has determined that IgH translocations occur often in myeloma, so that more than 90% of myeloma cell lines, 60% to 70% of advanced myeloma samples, and approximately half of MGUS tumors have at least one IgH translocation. Translocations involving the light-chain locus are less frequent, so that about 20% of advanced myeloma tumors and cell lines have IgG lambda translocations, but only a low percent of tumors and cell lines have Ig kappa translocations. Studies of translocation breakpoints have determined that each of five chromosomal loci are involved in 5% to 20% of myeloma tumors: 11q13 (cyclin D1), 4p16.3 (FGFR3+MM.SET), 16q23 (c-maf), 6p21 (cyclin D3), and 8q24 (c-myc). The translocations involving 4p16.3, 6p21, 11q13, and 16p23 have 14q32 breakpoints in or near the IgH switch regions or, less often, JH sequences consistent with errors in IgH switch recombination and somatic hypermutation. In contrast, translocation breakpoints involving 8q24 and other loci often have 14q32 breakpoints that are not near either IgH switch regions or JH sequences. Abnormal regulation of c-myc is important in BurkittÕs non-HodgkinÕs lymphoma and murine plasmacytomas.
Please see table on page 7.
Dr. Ho et al of the Royal Prince Alfred Hospital in Australia reported that they had studied myeloma cells from 21 patients (4 with stable MM, 17 with relapsing MM) and found translocations into the switch regions that were "illegitimate" (ie, not supposed to be there) in 57% (1 of 4 stable MM, 11/17 relapsing MM). Multiple switch regions were often involved in one tumor. These investigators then analyzed two of the genes that were most commonly involved in the translocations: 11p13 and 4p16. In the translocation 11;14, the cyclin D1 (CD1) gene could be transferred to 14q32 and the myeov gene, which remains on 11, can function abnormally because of the translocation. In the case of 4p16, the FGFR3 is transferred to 14q32, and MMSE is abnormally regulated on chromosome 4. They studied eight patients; 2 of them had CD1 expression, and one had increased myeov expression.
When disease course was examined, it was found that the presence of a switch translocation did not affect how long a patient lived (prognosis). Thus, the translocations cannot always be found, and they may not necessarily be the initial problem in the cell but may affect behavior of the cell later in the course of MM. The analysis of switch translocations is an important research tool; however, it is not currently helpful in treatment planning for an individual patient.
Chromosomal structure and stability is a critical factor in the pathogenesis of MM. The work of Drs Kuehl and Ho, along with others, is providing very useful insights into the role of these segments of chromosomes that are activated by movement of parts of one chromosome onto parts of others. This is helpful in understanding the underlying mechanisms of how myeloma is initiated and advances. It is likely that a better understanding of the genetic machinery of the myeloma cell will eventually lead to better methods of diagnosis and treatment.
A study of the cell-surface proteins on PCs can also explain the behavior of malignant PCs. Syndecan-1 (also known as CD138) is found on the surface of almost all myeloma cells, whether they are in the bone marrow or the blood. However, syndecan-1 is more than a marker of PCs; rather, it is a master regulator of molecular encounters. The myeloma cell can use syndecan-1 to attach itself to other myeloma cells or to other cells in the bone marrow. The syndecan-1 is connected to the inside of the cell to allow the cell to move around the bone marrow. Many myeloma investigators use beads with antibodies to syndecan-1 attached to help isolate PCs in the laboratory. Dr. Ralph SandersonÕs group at the Arkansas Cancer Research Center has shown that syndecan-1 is a binding partner to SP17, a heparan-binding protein originally thought to be found only in the testes. SP17 is found on the surface of myeloma cell lines and patient cells and can bind to syndecan-1 on the cell surface and promote myeloma cell-to-cell adhesion. The myeloma cell tends to organize the syndecan-1 at the leading edge of the cell referred to as the uropod. Other factors important to growth of PCs are also localized at the uropod. Syndecan-1 can be shed from the surface of the cell into the serum by the action of enzymes called proteases. Studies have shown that syndecan-1 accumulates in the serum of myeloma patients, and high levels of syndecan-1 in the serum correlate with a poor prognosis. Cells that express the soluble form of syndecan-1 are twice as invasive in the laboratory as cells that do not express any syndecan-1 at all. Syndecan-1 can also accumulate within the scarred area of bone marrow of patients with myeloma cells. This deposit of syndecan-1 could provide a reservoir for growth factors that are bound to syndecan-1, such as FGF-2.
FGFs are proteins that are capable of multiple functions. FGFs can stimulate some cells to grow and multiply, whereas in other situations FGFs may cause the cell to die or to stop growing. There are currently 22 known human FGFs that can interact with one or more of the four known FGF receptors (FGFRs) present on the cell surface. When an FGF binds to its FGFR, a signal is generated from outside the cell, through the FGFR to the inside of the cell. This sets off a cascade of reactions inside the cell that culminates in changes in the cell directed by the cell nucleus.
There are four reasons to study FGFs and FGFRs in MM:
1) Angiogenesis (new blood vessel formation) occurs in the marrow of patients with active MM but not in marrow from patients with MGUS. The knowledge that some FGFs are potent stimulators of endothelial cells (the cells that make up new blood vessels) provides rationale to learn whether myeloma cells make FGFs.
2) About 10% to 20% of patients with MM have a t(4;14)(p16;q32) translocation in the MM cells. The part of chromosome 4p16 that is moved to chromosome 14q32 contains the genetic code for FGFR3. Moving FGFR3 to the region of chromosome 14q32 (the area where the genetic code for antibody formation is contained) results in overexpression of FGFR3 in these cells. This may be one way to turn a PC malignant and suggests a way that FGFs binding to these FGFR3 receptors (the antenna-like structures that extend from the surface of the cell) may stimulate cell growth.
3) Syndecan-1 is a cell-surface receptor that can be identified by anti-CD138 antibody. Studies have shown that most PCs (benign and malignant) have syndecan-1. Since syndecan-1 binds FGF, this association provides a clue that FGF may play a role in MM.
4) Thalidomide is an important new treatment for MM. Understanding how thalidomide works will provide insight into other ways to kill myeloma cells. We know that if a pregnant woman takes thalidomide it can interfere with limb development in the fetus. It
turns out that FGFs are critical components of the cell machinery used to form limbs in a developing fetus. This is a hint that perhaps thalidomide works in MM by interfering with FGF/FGFR interaction.
A first step in understanding the role of FGFs and FGFRs in MM is to determine whether myeloma cells make FGFs/FGFRs. Human myeloma cell lines are made from myeloma cells from patients with the disease. The cells are placed in culture in the laboratory and can be grown and used for experiments for many years. They are very useful, because the investigator can perform many experiments without having to go back and ask the patient for another blood or marrow sample. Dr. Witzig and colleagues studied myeloma cells for production of FGFs and FGFRs. Most of the cell lines made FGF-2, and some made FGF-5 and FGF-9. The cell lines did not make FGF-1, -3, -4, -6, or -7. There are many other FGFs, but they have not yet been studied and less information is known about them.
Studying the cell lines for FGFRs is also important, because if the myeloma cell makes FGFR and FGF, it is possible that a myeloma cell can stimulate itself (referred to as autocrine stimulation). All the cell lines were shown to make at least one FGFR, and synthesis of all four FGFRs was found. Myeloma cells were isolated from the bone marrow of eight patients with MM. We found that all samples contained the message for FGF-2 production, one patientÕs cells had message for FGF-5, and one had FGF-9. Three patients had FGFR-1, and two each had FGFR-2, -3, and 4. Dr. Keith Stewart of the Princess Margaret Hospital in Canada recently demonstrated that when bone marrow cells that express FGFR-3 were transplanted into mice, the mice developed cancer of the B cells. This indicates that the abnormal expression of FGFR-3 on myeloma cells can be oncogenic. In other words, this can cause the cells to become malignant. If a myeloma cell has the t(4;14)(p16;q32), the FGFR-3 is placed into a region very important to the function of B-cells. This explains one way a PC can become malignant. It is clear that most cases of MM do not have the t(4;14)(p16;q32); thus, there appear to be multiple genetic pathways to produce MM . It is not clear what the relationship is between the genetic abnormalities and response to treatment or prognosis. This work indicates that myeloma cells can make FGFs and FGFRs. What function these cytokines have in myeloma cells is yet to be explained, but they may contribute to the increased angiogenesis found in the marrow of patients with MM.