Effects of EDTA on Routine and Specialized Coagulation Testing

BiologyEffects of and EDTA Bioinformatics on Routine of and Myeloma Specialized Cell SaeidCoagulation Abroun,1 Najmaldin Testing Saki, and2 Rahim an Fakher, Easy3 FarahnazMethod Asghari to Distinguish4

EDTA-Treated1Department of Hematology and Bloodfrom Banking, Citrated School of Medical Sciences,Plasma Tarbiat ModaresSamples University, Tehran, Iran; 2Research Center of Thalassemia & Hemoglobinopathy, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran; 3Physiology Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran; 4Faculty of Medicine, University Rondaof Rostock, A.Rostock, Crist, Germany1,2 Kathie Gibbs,2 George M. Rodgers,1-4 Kristi J. Smock1,2,4 Received May 14, 2011; December 13, 2011, February 10, 2012; August 7, 2012 1ARUP Institute for Clinical and Experimental Pathology, Salt Lake City; 2ARUP Hemostasis/Thrombosis Laboratory, ⌦Salt Lake City; Departments of 3Medicine and 4Pathology, University of Utah, Salt Lake City, Utah, USA Received October 7, 2009; received in revised form November 5, 2009; accepted November 6, 2009 ABSTRACT approximately 14,600 new cases of myeloma are diagnosed ABSTRACT each year in the United States and that it develops in 1 to Multiple myeloma (MM) is a plasma cell disorder that 4 people per 100,000 people each year. With conventional occurs in about 10% of all hematologic cancers. The major- treatment, the prognosis is 3 to 4 years, which may be Coagulation testing is performed with citrate-treated ity of patients (99%) are over 50 years of age when diag- treatedextended with to 5citrate, to 7 yearswhich with anticoagulates advanced treatments. by chelating MM ⌦cal is- plasma. Samples submitted in other anticoagulants, such nosed. In the bone marrow (BM), stromal and hematopoi- ciumthe second ions necessarymost prevalent for calcium-dependent blood cancer (10%) coagulation after non- as EDTA, should not be tested. We aimed to evaluate the etic stem cells (HSCs) are responsible for the production of reactionsHodgkin’s [1,2]. lymphoma Calcium (NHL). is replenished It represents by the approximately test reagents effects of EDTA on routine and specialized coagulation blood cells. Therefore any destruction or/and changes within at1% the of timeall cancers of testing. and 2%EDTA of all is canceran anticoagulant deaths. Although that che its- tests and to establish sodium tetraphenylborate testing as a the BM undesirably impacts a wide range of hematopoiesis, latespeak calciumage of onset irreversibly, is from 65a quality to 70 yearsthat hasof age, the recentpotential statis to- quick and reliable method to identify EDTA-treated plasma causing diseases and influencing patient survival. In order dramaticallytics indicate bothaffect an the increasing results and incidence interpretation and an earlierof coagu age- samples. We performed the following measurements on to establish an effective therapeutic strategy, recognition lationof onset tests [1]. [3-8].Usually, EDTA MM issamples diagnosed must by bloodbe identified sample tests and citrate- and EDTA-treated plasma samples from 10 healthy of the biology and evaluation of bioinformatics models for rejectedincluding by peripheral the coagulation blood smear laboratory and protein to avoid electrophore negative- volunteers: sodium tetraphenylborate testing, prothrombin myeloma cells are necessary to assist in determining suitable impactssis, but inon somepatient cases care. microscopic In the present examination study, we of aimed the BM to time (PT), partial thromboplastin time (PTT), potassium methods to cure or prevent disease complications in patients. evaluatebiopsy and the x-rays effects of bonesof EDTA were alsoon routinegood indicators and specialized of MM. concentration, and functional assays for factors II, V, VII, This review presents the evaluation of molecular and cellular coagulationIn 2003, the tests International and to describe Myeloma a new Working method Group for identify agreed- VIII, IX, X, XI, XII, proteins C and S, and antithrom- aspects of MM such as genetic translocation, genetic analysis, ingon aEDTA-treated diagnostic criteria plasma for samples. symptomatic and asymptomatic bin. Mean values for citrate- and EDTA-treated plasma cell surface marker, transcription factors, and chemokine myeloma,In our referencewhich was laboratory subsequently setting, updated we do in not2009: receive The were most different for PT, PTT, factors V and VIII, and signaling pathways. It also briefly reviews some of the mecha- ⌦presenceprimary-draw of unexplained tubes, making anemia it difficultby hemoglobin to discern < 10 whether g/dL, proteins C and S. Sodium tetraphenylborate testing cor- nisms involved in MM in order to develop a better under- thekidney correct dysfunction anticoagulant follow wasby raisedused. serumAs part creatinine, of our quality- a high rectly classified 100% of citrate-treated and EDTA-treated standing for use in future studies. ⌦erythrocyteassurance plan, sedimentation we measure rate, the and prothrombin a high serum time protein (PT) samples. We confirm that EDTA has effects on coagula- and> 30g/L, the partial which thromboplastin especially raised time immunoglobulin (PTT) on all samples. (Ig) in tion assays. Sodium tetraphenylborate testing is a quick, KEY WORDS: Multiple myeloma◆◆•◆◆Chemokines◆◆•◆◆ Thisserum process and/or highallows proteinurea, us to detect which erroneously indicate Bence submitted Jones simple, and inexpensive method for coagulation laborato- serumprotein samples,in urine, whichhyper calcemiahave markedly > 2.75 prolongedmmol/L, infection, PT and ries to identify samplesSignaling erroneously pathways◆◆•◆◆Bioinformatics submitted in EDTA. PTTbone valuespain by (often lytic lesions>150 seconds) or osteoporosis but correct with to compression within the Lab‑Hematol. 2009;15:45-48. normalfractures, range neurological in a 1:1 mixingsymptom study. by hypercalcemia,This approach ordoes spinal not INTRODUCTION reliablycord damage identify by backEDTA-containing bone fracture. plasmaMM can samples, also be which diag- KEY WORDS: Citrate◆◆•◆◆EDTA◆◆•◆◆Coagulation usuallynosed if demonstratethe population prolonged of plasma (but cells measurable)in the BM biopsy PT and are Multiple myelomatesting◆◆•◆◆Preanalytical (MM) is a B-lineage cell variables malignancy PTT> 10% values [2,3]. that Based often on Internationaldo not correct Staging in a 1:1 System mixture (ISS) [9]. or characterized by the monoclonal expansion of malignant SuspectedDurie-Salmon EDTA-containing staging guidance, plasma there canare 3be classified confirmed stages by plasma cells within the bone marrow (BM). There are undetectabilityof MM (Table) of[4,5]. calcium (due to avid calcium chelation by INTRODUCTIONapproximately 45,000 people in the United States diagnosed EDTA)It is andlikely increased that MM potassium has evolved values from(EDTA pre-malignant tubes usually with MM, and the American Cancer Society estimates that containstages of EDTA clonal asplasma a potassium cell proliferation salt) [1,6,8-10]. termed Conducting monoclo- Coagulation testing is performed on plasma samples thesenal gammopathy chemical tests of is undetermineddisruptive to our significance laboratory (MGUS).work flow Correspondence and reprint requests: Saeid Abroun, Department of becausePatients theymay arebe notdiagnosed available with at theMGUS coagulation if they bench. fulfill We the HematologyCorrespondence and and Blood reprint Banking, requests: School Kristi of J.Medical Smock, Sciences,ARUP Labo Tar- havefollowing found 3 thatcriteria: testing 1) patientA monoclonal samples withparaprotein sodium tetraband- biatratories, Modares 500 ChipetaUniversity, Way, Tehran, Salt LakeIran; City,+98 21UT 82883860; 84108, USA fax: (e-mail:+98 21 phenylborateless than 30 g/Lis a simple(< 3 g/dL); and reliable 2) Plasma way cellsto identify less than EDTA- 10% [email protected]). (e-mail: [email protected]). of bone marrow cell population; and 3) No evidence of

27 28 Abroun et al

is as yet unknown. B-cells originate from hematopoietic stem TABLE. Multiple Myeloma Staging System* cells (HSCs) in BM. Progenitor B-cells (B220+, CD19−, + International Staging CD117 ), B220 in mice, known as CD45 in humans, + + + + Stage System (ISS) Durie-Salmon and precursor B-cells (B220 , CD19 , CD25 , CD38 ) All of: Hb > 10 g/dL, normal calcium; are derived from HSCs and finally differentiated to mature + + + − β2-microglobulin skeletal survey: normal or single plas- B-cells (CD19 , CD20 , CD24 , CD93 ). Plasma cells + − − + + + + I (β2M) < 3.5 mg/L and macytoma or osteoporosis, serum (CD9 , CD19 CD24 , CD38 , FcγRII , CXCR4 , Blimp1 albumin ≥ 3.5 g/dL paraprotein level IgG < 5 g/dL, IgA < 3 [B-lymphocyte–induced maturation protein 1]) derived from g/dL, urinary light chain excretion activated B-cells [12-14]. Myeloma cells are the transformed β2M < 3.5 and albumin counterparts of post–germinal-center BM plasma blasts or II Fulfilling the criteria of neither I nor III < 3.5 plasma cells [15]. In silico tools offer an attractive alternative One or more of: Hb < 8.5 g/dL, high strategy to the cumbersome experimental approaches [16]. calcium > 12 mg/dL; skeletal survey: These computational tools have metamorphosed over the III β2M ≥ 5.5 mg/L three or more lytic bone lesions, years into complex algorithms that attempt to efficiently pre- serum paraprotein IgG > 7 g/dL, IgA dict the binding of peptides to receptors. > 5 g/dL *Stages I, II, and III of the Durie-Salmon staging system can be divided GENETIC TRANSLOCATION into A or B depending on serum creatinine: A, serum creatinine < 2 mg/ dL (< 177 mol/L), and B, serum creatinine > 2 mg/dL (> 177 mol/L). Chromosomal disorders existed in 20% to 60% of the patients in whom MM was detected early and in 60% to 70% of the patients with the advanced form of the disease bone lesions, anemia, hypercalcemia, or renal insufficiency [13]. The translocations observed in MM patients mostly related to the paraprotein. Although presence of monoclo- involved the immunoglobulin heavy (IGH) chain locus nal paraprotein is 1 of the symptoms of MGUS, several on chromosome 14q32 [15]. Common genetic changes other disorders can also present a monoclonal gammopathy, that were observed included chromosome 1 abnormality, such as AIDS, chronic lymphocytic leukemia (CLL), NHL, 11q deletion, t(11;14), t(4;14), trisomy 12, 13q deletion, Hepatitis C, Lupus, Waldenstrom macroglobulinemia, and monosomy 13, IGH rearrangement, 17p13 deletion, and Guillain-Barré syndrome [6]. Therefore the percentage of hyperdiploidy [17]. The t(4;14), t(14;16), and t(14;20) are plasma cells in BM to confirm MGUS has been a critical highly associated with a deletion of chromosome 13 in MM diagnostic tool. MGUS occurred in 3% of people older patients [18]. The t(11;14)(q13;32) is the most prevalent than 50 years and after follow-up of the surviving patients, translocation in MM with a prevalence of 15% to 20%. we found that it switched into MM or a similar myelopro- This translocation causes a disruption in the expression of liferative disorder at the rate of about 1% to 2% a year, or cyclin D1, which in turn leads to an increase in the level 17%, 34%, and 39% at 10, 20, and 25 years, respectively. of cyclin D1 expression. Cyclin D1 usually promotes the However, because most patients with MGUS were elderly, progression of the cell from the G1 growth arrest phase they died of something else and did not go on to develop into the S phase in the cell cycle division. Because of this MM. When this was taken into account, only 11.2% devel- translocation, the cell cycle is accelerated leading to high oped myeloproliferative disorders [7,8]. Some patients with rates of cell proliferation. The plasma cells don’t usually MGUS show few if any symptoms that are usually missed proliferate, but because of a chromosomal abnormality, the by the physician. These patients therefore do not receive plasma cells have an accelerated cell cycle rater leading to a follow-up, and the physician only notices their illness when neoplastic proliferation [19,20]. Expressing CD19 or CD20 it has developed into MM. Therefore the percentage of MM is associated with the morphology of small mature plasma patients with a MGUS background is probably more than cells, but in MM cells, with the occurrence of t(11;14), the reported. The potency of MGUS causes MM malignancy, expression of CD19 and CD20 become mutually exclusive. hence it is possible that all MM patients have a background One way of distinguishing between normal plasma cells and of MGUS that has been noticed in some patients but not in myeloma cells is that normal plasma cells express CD19 others. The lesion in MGUS is in fact very similar to that in whereas myeloma cells don’t. However, in Walden storm’s MM. There is a predominance of clonal plasma cells in the cells both CD19 and CD20 markers are expressed. MM BM with an abnormal immunophenotype (CD38+, CD56+, patients with CD19−, CD20−, and CD27− are character- CD19−) mixed in with cells of a normal phenotype (CD38+, ized by the detection of frequent t(4;14) or t(14;16) and CD56−, CD19+) [9,10]; in MGUS, on average, more than fewer than normal diploid chromosomes [21]. The t(4;14) 3% of the clonal plasma cells have the normal pheno- (p16;q32) occurs in 15% of MM patients, and it is associ- type, but in MM less than 3% of the cells have the normal ated with a poor prognosis. The t(4;14) up-regulate fibro- phenotype [11]. What causes MGUS to transform into MM blast growth factor receptor 3 and multiple myeloma SET Myeloma and Bone Lesion 29 domain (MMSET) genes. This translocation leads to the interleukin (IL)-10 production. This transcription factor disruption of transforming acidic coiled–coil-containing (also known as the T-cell transcription factor) is expressed protein 3 (TACC3) expression; although the function of the in monocytes and macrophages. In addition to IL-6, IL-10 TACC3 gene has not yet been determined, it is speculated is considered as 1 of the most important cytokines regulat- that it may be involved in cell growth and differentiation. ing the proliferation and cellular characteristics of myeloma Expression of TACC3 gene is up-regulated in some cancer cells [20,25,26]. The t(6;14)(p21;q32) involving cyclin D3 cell lines, and in embryonic day 15 in mice. Disorders of pathway occurs in 3% of myeloma cases. In situ hybridization chromosome 13 in MM patients relates to a poor prog- methods have shown the prevalence of cyclin D1 over-expres- nosis of the disease [20,22,23]. Previous studies have sug- sion to be 32%. Cyclins D1, D2, and D3 (CCND1, 2, 3) are gested that inhibition of receptor tyrosine kinas (RTK) has regulated by proteasomal degradation. Their overexpression been essential in disrupting mitogen activated protein kinase in MM has a prognostic value [27,28]. The t(6;14)(p25;q32) (MAPK) proliferation and survival signals in t(4;14) MM led to the juxtaposition of the heavy chain locus in multiple [24]. The t(14;16)(q32.3;q23) has been detected in 5% to myeloma oncogene 1 (MUM1) to the interferon regulatory 10% of MM patients and leads to a deregulation of C-maf factor 4 (IRF4) gene. The t(9;14)(p13;q32) translocation is proto-oncogene. C-maf is a basic zipper transcription factor rarely found in patients with myeloma and leads to a deregu- that is involved in many basic cellular processes including lation of the PAX-5 gene [20].

FIGURE 1. Representation of cellular interactions in the bone marrow (BM) of multiple myeloma (MM) patients. Normally, receptor activator of NF-kappa B (RANK) is expressed by osteoclasts, and its ligand (RANKL) is represented by osteoblasts; this connection leads to activation of some signaling pathways in certain cells. Osteoprotegerin (OPG) is released by osteoblasts and bound to RANKL competi- tively and in this way limits interaction between RANK and RANKL. Myeloma cells via expressing RANKL and connecting to RANK and eliminating of environment OPG (by expressing its decay receptor ie, CD138) take part in more activation of osteoclast and in other hands by expressing some inhibitor factors such as sFRP-2, inhibit the Wnt/β-cat signaling pathway in osteoblasts. 30 Abroun et al

ROLE OF MYELOMA CELLS IN BONE LESIONS stromal cells of BM and osteoblasts; it attaches to RANKL and inhibits its attachment to RANK. The interaction The BM microenvironment is full of stem cells, progeni- between RANKL and RANK is important for osteoclast tor, precursor, and differentiated cells with different linage formation and activity. The inhibition of RANKL by OPG is and functions. In addition to hematopoietic cells, other leads to the prohibition of bone resorption [40]. New patho- cells such as fibroblasts, adipocytes, endothelial cells, osteogenic studies on myeloma cells show that malignant plasma blasts, osteocytes, and osteoclasts are present in the BM. In cells can differentiate in the BM microenvironment to osteo- a healthy body, there is a balance between osteoclast and clast-like cells and directly take part in bone resorption [39]. osteoblast activity, but in MM this balance is disrupted [29]. It is indicated that in MM, osteoblast apoptosis caused by Furthermore, in 90% of the MM patients the osteolysis leads their permanent exposure to high concentrations of TNF-α, to bone destruction [30]. It has recently been reported that INF-γ, IL-1β, IL-6, and ICAM1/LFA1 molecules [30]. osteoprotegerin (OPG), receptor activator of NF-kappaB We analyzed the potential of myeloma cells to induce cord ligand (RANKL), and receptor activator of NF-kappa B blood HSC differentiation. The results showed an increased (RANK) systems are the most important regulators for bone expression of myeloid and monocytoid markers in the co- formation [31]. Decreasing OPG levels and an increasing culturing of myeloma cells with HSCs. We suggest that the RANKL to OPG ratio leads to the activation of osteoclasts presence of myeloma cells in BM may play an essential role and ultimately bone destruction. Recently, 3 main groups of in HSC differentiation to monocytoid (osteoclastic) lineage factors considered as main osteoclast inducers in MM have (Figure 1) [41]. been detected: RANKL, macrophage inflammatory protein (MIP)-1α, β, and stromal cell–derived factor (SDF)-1α CYTOKINES [32]. The RANK ligand is part of the tumor necrosis factor gene family and is a major osteoclastogenic factor involved BM is a suitable microenvironment not only for the in MM bone disease [33]. Reverse transcription polymerase proliferation and survival of myeloma and hematopoietic chain reaction (RT-PCR) analysis of the human myeloma malignant cells, but also for many solid tumors such as cell line U-266 and primary myeloma cells from myeloma prostate or breast cancer cells [20]. Myeloma cells secrete patients demonstrated that these cells expressed RANK and potential proangiogenic cytokines such as vascular endothe- RANKL at mRNA levels and can induce myeloma cell auto lial growth factors (VEGF), basic fibroblast growth factor activation [34]. Circulating serum tRANKL was significantly (bFGF), Angiopoietin-1 (Ang-1), and osteopontin (OPN), elevated in patients with MM [35]. tRANKL can bind to its and this is the cause of increasing in BM microvessel density specific receptor (RANK) on the myeloma cell surface and in such patients [20,22]. Myeloma cell lines express a high induce some signals inside the cell. It has been clarified that level of chemokine receptors such as CXCR3, CXCR4, IL-6 and its specific receptor (CD126) exist at an increased CCR1, CCR5, and CCR6. Myeloma cells express SDF-1α level in the serum of MM patients. IL-6 is produced by BM protein and its receptor (CXCR4) at different levels. The stromal cells (BMSCs), and it is a growth factor for osteo- growth of myeloma cells is mediated by the autocrine and clasts and myeloma cells, which promotes their proliferation paracrine secretion of IL-6. Furthermore, IL-6 production and antiapoptotic processes. The main effect of IL-6 is to from myeloma cells is enhanced by IL-13 or CD40 stimu- increase osteoclasts precursor in BM [36,37]. It is indicated lation, leading to accelerated autocrine secretion and more that the number and activity of osteoblasts are decreased in proliferation and growth. IL-6 also has an anti-apoptotic the adjacent surfaces of myeloma cells. The Wnt pathway is effect on myeloma cells and thus, IL-6 supports the expan- the main signaling pathway in osteoblasts, and suppressing sion of myeloma cells by both stimulating cell proliferation this pathway leads to a decrease in the function of osteoblasts and preventing apoptosis [42-45]. The paracrine production [32]. It had been shown that BM mesenchymal stem cells are of IL-6 by stromal cells in BM induced signals, which are abnormal in MM, which can accelerate the proliferation of potential osteoclastogenics and function as osteoclast pre- MOLP-6 myeloma cell line rather than the mesenchymal, cursors [40]. Osteoblasts secrete hematopoietic growth fac- which is what is obtained from healthy donors [38]. Disabil- tors, such as hepatocyte growth factor, IL-6, FGF, and also ity of osteoblasts in lesion renovation is caused by induced TGF-β, which has a role in bone remodeling [46]. Myeloma mechanisms by malignant plasma cells such as direct cytotox- cells can be classified into at least 5 subpopulations: ity against osteoblasts by Fas-L and expressed tumor necrosis MPC-1−, CD45+, CD49e− (immature), MPC-1−, CD45−, factor (TNF) receptors by myeloma cells induce apoptosis in CD49e− (immature), MPC-1+, CD45−, CD49e− (intermedi- them [39]. RANKL is expressed by osteoblasts and stromal ate), MPC-1+, CD45+, CD49e− (intermediate), MPC-1+, cells of BM and the interaction between RANKL and its CD45+, and CD49e+ (mature). Among these cells only receptor (RANK), which is expressed in osteoclasts. RANKL MPC-1–, CD45+, and CD49e− immature myeloma cells can functions by stimulating osteoclast formation and bone respond directly to IL-6 to proliferate [47]. IL-6 and IGF-1 resorption. OPG is a soluble receptor that is secreted by are the main myeloma growth factors and heparin-binding Myeloma and Bone Lesion 31 epidermal growth factor (HB-EGF)–mediated survival of expression of CD28 by myeloma cells is associated with human myeloma cell lines is dependent on the presence of myeloma metastase, IL-8 directly stimulates the differen- an autocrine IGF-1 loop, whereas the activity of recombi- tiation of human peripheral mononuclear cells to osteoclasts. nant APRIL is not [44,48]. Usually in non-malignant cells, IL-6 (the main myeloma growth factor) can increase the pro- IGF-I receptors were phosphorylated by its ligand IGF-I, duction of MCP-1 in MM [42]. but it had been shown that in some myeloma cell lines, with a high expression of IL-6R, Il-6 activates janus kinase (JAK)- CD MARKERS AND EXPRESSED MOLECULE signal transducer and activator of transcription (STAT) and MAPK pathways that support cell proliferation. Fur- CD81, or target of anti-proliferative antibody–1 (TAPA- thermore, it can induce IGF-IR phosphorylation by recep- 1), has a very broad cellular distribution, being expressed on tor–receptor interaction and IL-6 induced unspecific PI3-K T- and B-lymphocytes. CD19, Leu13, and CR2 (CD21) pathway followed by the activation of AKT signals inside are all a B-cell surface signal transduction complex [54]. the myeloma cell, which can in turn support survival and CD19 is known as a malignant CD marker, and its absence prevent apoptosis of cells. In this situation, IL-6 plays a piv- is observed in almost all MM patients. Malignant plasma otal role in the growth and survival of in vitro myeloma cell cells (myeloma cells) isolated from the MM patients lack lines and myeloma cells isolated from MM patients [44]. any CD19 expression; non-malignant plasma cells isolated IGF-1 could increase the mitosis of MM cell lines and can from the healthy donors show expression of CD19 antigens. serve as a chemo attractant for MM cells through the PI3K Although expression of CD19 is a key to distinguishing nor- pathway [49]. IL-6 overproduction leads to the activation of mal plasma cells from malignant plasma cells, it is intriguing osteoclasts, induction of MMPs and VEGF, expression of an that sometimes both CD19– and CD19+ plasma cells exist adhesion molecule, and induction of Th17 cells and hyper γ in pre-myeloma states, including MGUS [21]. The in vitro globulinemia [50]. studies have shown that proliferation of few CD19+ myeloma IL-3 may play a role in the regulation of bone formation cell lines is slower than CD19–. Therefore, it is possible that in myeloma bone disease. BM plasma cells from myeloma CD19 has a tumor suppressor activity in MM [55]. patients were found to inhibit osteoblast differentiation, and CD27 is a memory marker because its expression is this effect could be blocked by a neutralizing antibody for restricted to germinal center cells, memory cells, and plasma IL-3 [40]. IL-3 and IL-7 are potential inhibitors of osteo- cells. Gene expression profile studies of normal plasma cells blast differentiation in MM [33]. Hepatocyte growth factor and myeloma cells have identified CD27 as being 1 of the (HGF) is produced by myeloma cells and is increased in the most significant genes lost by myeloma cells. However, half serum of MM patients. Increasing levels of HGF are corre- of MM patients retain CD27, and its expression is associ- lated with a poor prognosis. Serum concentrations of HGF ated with a better prognosis. CD27 expression is lost with were negatively correlated with levels of bone–specific alka- the progression of myeloma. CD28, a T-cell–specific marker, line phosphatase in MM patients, suggesting a role for HGF is aberrantly expressed by primary myeloma cells but not in myeloma bone disease. HGF was found to inhibit bone by normal plasma cells. Myeloma cells express 1 co-receptor morphogenetic protein (BMP)-induced osteoblastogenesis, of CD28, CD86, but not the other, CD80 [21]. Human including expression of the transcription factors Runx2 and myeloma cells from about 10% of cases with MM expressed Osterix, potentially by inhibiting nuclear translocation of CD33, a myeloid marker, and had monocytoid morphol- receptor-activated SMADs [44]. It was reported that Dick- ogy with convoluted nuclei. IL-6 regulates the expression of kopf (DKK) increased the formation rate of osteoclasts and CD33 in a number of BM myeloma cells. High IL-6 concen- bone resorption. The systemic administration of anti-DKK1 tration leads to a decreased expression of CD33 on myeloma antibodies was shown to block osteoclast formation in vivo cells. IL-6 induces the activation of MYC, and joining MYC in a murine model of myeloma bone disease [51]. Myeloma to CCAAT-enhancer-binding protein (CEBP) gene promoter plasma cells secrete Wnt's signaling inhibitor by DKK1 [52]. suppressed the expression of the CEBP gene and ultimately BM plasma cells of patients with MM expressed VEGF, led to the down-regulation of CD33 gene expression. A whereas both VEGFR-1 and VEGFR-2 were considerably JAK-STAT signaling—responsible for the decrease in CD33 increased in the BMSCs, which suggested that a paracrine expression—is activated by IL-6 [56]. The presence of hyal- growth pathway mediated by VEGF-activated stromal cells, uronic acid (HA) in the environment of MM cells increases secretion of IL-6 on their part, and subsequent activation their survival potential and proliferation, and this can be of plasma cells may occur. In active MM patients, VEGF- because of HA interactions with HA-binding receptors A isoform is produced by plasma cells over expression of (CD44) on MM cells [30]. There is evidence that CD44 can VEGFR-2 by BM microvessels and expression of VEGFR-1 also bind to OPN [57]. In an in vitro angiogenesis assay sys- by the other stromal cells, secretion of VEGF- C and VEGF- tem, it was demonstrated that conditioned media from MM D by stromal cells, and expression of their cognate recep- cells has a proangiogenic effect, and OPN has a critical effect tor VEGFR-3 by plasma cells [53]. Since the abnormal on this outcome [58]. 32 Abroun et al

The expression of CD45 on human myeloma cells has leukocyte trafficking, and pathogen virulence. Therefore, the been reported on immature myeloma cells but not on mature change in sight of syndecan-1 from the cell surface to the elements. Although the CD45 molecule is expressed on the extracellular compartment has important pathological conse- cell surface of all hematopoietic cells, it has been known as quences [65]. CD138 allows excellent assessment of plasma a pan-leukocyte antigen, and plasma cells also expressed it, cell numbers and distribution in BM biopsies [66]. but its expression in primary myeloma cells and myeloma CD221 is also called IGF-1R. Lack of CD27 or CD45 cell lines has been shown to be quite variable. Most myeloma or over-expression of CD221 has been shown to be associ- cells do not express CD45, but a few MCP-1– immature ated with a more adverse prognosis. CD221 is aberrantly and some MCP-1+ cells do express CD45. Without IL-6 in expressed in myeloma cells when compared to normal and the culture, those MCP-1– CD45+ myeloma cells undergo reactive plasma cells. Patients with a high level of IGF-1R apoptosis more easily than MCP-1– CD45– cells, suggesting expression have a shorter survival period. Over-expression of that the MCP-1– CD45+ myeloma cells essentially need IL-6 IGF-1R is associated with t(4;14) and lack of CD45 expres- [59]. Engagement of B-cell antigen receptor by anti-IgM sion [21]. IGF-1 is an important survival and growth factor antibodies induces the activation and recruitment of src fam- in multiple myeloma [44,67]. ily protein tyrosine kinases (PTKs), such as Lyn, Fyn, and Blk. B-cells lacking CD45 are incapable of proliferating in TRANSCRIPTION FACTOR–MEDIATED SURVIVAL response to anti-IgM stimulation. Furthermore, T-cell matu- SIGNAL PATHWAYS ration in the thymus is impaired in CD45-deficient mice. These defects in lymphocytes are though to occur because of Pathways or central mediators that are effective in pro- their inability to activate the src family PTKs [60]. Although liferation and survival of MM cells involve phosphatidyl/ the CD45 molecule is expressed on B-cells and plasma cells, inositol3-kinase/Akt (PI3K/Akt), Notch, wingless (Wnt), most myeloma cells lose their expression. Moreover, CD45 and nuclear factor-kappa B (NF-κB). Like the Ras/MAPK expressing isoforms are different between normal plasma pathway, the PI3K/Akt pathway might be activated by vari- cells and myeloma cells. Myeloma cells express CD45RO ous BM-derived cytokines and chemokines [68]. but not CD45RA, but plasma cells express CD45RA but not Stimulation of Wnt signaling enhances proliferation of CD45RO [59]. MM cells, whereas its inhibition attenuates growth of MM cell lines in vitro. Wnt signaling is a negative regulator of CELLULAR ADHESION MOLECULE early B-lymphopoesis [69] and has a vital role in bone mass regulation and stem cell self-renewal [40,70]. The majority of CD18, CD29, CD54, and CD56 are expressed by MM cells express Wnt3 factor, which was shown to activate myeloma plasma cells [61]. A wide range of integrins are the Wnt3/RhoA/ROCK pathway in BMSCs. In turn, this expressed by MM cell lines (ie, β2 integrins). α4β1 integ- was revealed to have an anti-apoptotic influence on adher- rin (VLA4) is over-expressed in drug resistant MM cells. ing MM cells [30]. The Wnt signaling pathway has been MM cells express CXCR4, and the SDF-1/CXCR4 axis is shown to play a key role in osteoblast differentiation and a key regulator of MM cell homing, adhesion, and motil- has been implicated in osteoblast suppression in myeloma ity [62,63]. CD138 (Syndecan-1) expression is restricted [32]. 1,25(OH)2D3 and calcium modulate cell growth to both plasma cells and myeloma cells. Human CD38, a because the activated signaling pathway via the vitamin D pleiotropic molecule with ADP-ribosyl cyclase activity, regu- receptor (VDR) and calcium-sensing receptor (CaR), affects lates activation and growth of several cell types. All normal the Wnt pathway. CaR activation leads to Wnt/β-catenin and malignant plasma cells expressed CD38 and CD138. pathway inhibition. It has been shown that 1,25(OH)2D3 Some of the MM cell lines are not myeloma cell lines but inhibits growth and promotes differentiation of human colon B-cell lines transformed by Epstein-Barr virus [21]. Syn- adenoma and carcinoma cells by inhibiting the up-regulation decan-1 is also highly expressed by myeloma cells; it is a low of cyclin D1 expression, a key element in cell cycle control affinity receptor of bFGF, which is a modulator of myeloma [71]. Patients with the highest vitamin D levels had the low- cell growth and survival [64]. Heparinase can also enhance est risk for osteoporosis [72]. Although Wnt signaling activa- release of the syndecan-1 heparan sulfate proteoglycan from tion has been linked to several forms of cancer, increasing the surface of tumor cells. This occurs through heparan- Wnt signaling in the BM, through direct administration of ase-mediated up-regulation of extracellular signal-regulated Wnt3a, can inhibit bone destruction and MM tumor growth kinase (ERK) phosphorylation leading to enhanced expres- in vivo. MM plasma cells secrete the Wnt-signaling inhibi- sion of matrix metal proteinase (MMP-9) syndecan-1 shed- tor DKK1 [52]. Secretion of Wnt inhibitor factors (sFRP-2) ders on cell surface. The enhanced shedding of syndecan-1 by myeloma cells is 1 factor in myeloma bone lesion. As is important biologically because the released proteogly- previously mentioned, the Wnt-signaling pathway regulates can remains active and can influence vast functions such proliferation, survival, and functional half-life of osteoblasts, as tumor growth and metastasis, chemokine localization, while on the other hand, Wnt down-regulation induces Myeloma and Bone Lesion 33 apoptosis in them [69,73]. Wnt/Ca2+ is another pathway MM cells to inhibit Runx2 activity in human mesenchymal that can be activated via connection between some Wnts and stem and osteoprogenitor cells [77]. its specific receptor (Fz) and that can stimulate intracellular 2+ Ca release from endoplastic reticulum (ER). This pathway B-CELL DEVELOPMENT IS REGULATED BY is dependent on G-proteins. The calcium release and intra- CORE PROTEIN PAX-5/EBF cellular accumulation activate several Ca2+-sensitive proteins, including protein kinase C (PKC) and calcium/calmodulin- CD19+ B-cells and plasma cells both express Pax-5 mRNA dependent kinase II (CamKII) [74]. and CD19 [54]. A transcription factor, B-cell–specific activa- Notch signaling seems to strongly affect functions of tor protein (BSAP), encoded by the Pax-5 gene, plays a crucial both MM cells and BMSCs. MM cells express Notch-1 and role in the expression of the CD19 gene. The Pax-5 gene Notch-2 and Jagged-1 and Jagged-2, and BMSCs express belongs to the family of paired box containing genes (Pax), Notch-1, Jagged-1; consequently, the Notch signaling path- and the Pax genes in mammals have been implicated in the way was found to be activated in both BMSCs and MM regulation of morphogenesis and pattern formation. In mice cells [30]. lacking Pax-5/BSAP, the B-cell development was arrested at XIAP is the most potent direct endogenous caspase the stage of pro–B-cells without expressing CD19. Neither inhibitor and is considered a key factor in controlling the CD19 nor Pax-5 mRNA and proteins can be detected in apoptotic threshold in cancer cells. Myeloma cells have myeloma cells and cell lines, whereas non-malignant plasma been shown to express high levels of XIAP protein that were cells express both CD19 and Pax-5 mRNA. These finding tightly regulated during growth factor stimulation or stress imply that the altered expression of Pax-5 is responsible for conditions [75]. the loss of CD19 expression in human MM [55]. IL-6 receptor complexes consist of IL-6Rα and gp130, the latter being responsible for signal transduction. Upon IN SILICO AND BIOINFORMATICS APPROACHES IL-6 binding to the IL-6Rα, the gp130 forms dimmers followed by activation of tyrosine kinases JAK1 or Tyk2 con- Genetic Translocation and Genomic Analysis stitutively associated with the gp130. The activated JAKs There are a variety of in silico and bioinformatics tools phosphorylate 6 tyrosine residues located in the cytoplasmic and methods that are applicable and helpful to discover region of the gp130-recruiting signal transducers and activa- candidate oncogenic chromosomal aberrations. One of these tors of transcription STAT3, whose tyrosine residues are also novel bioinformatics approaches applies assuming outlier phosphorylated by JAKs. The phosphorylated STAT3 forms gene expression called COPA (cancer outlier profile analy- dimmers and translocates to the nucleus (JAK-STAT path- sis) or ONCOMINE (a cancer microarray database and way). Another major signal transduction pathway via gp130 integrated data-mining platform), which, in general, offers a is the Ras-MAPK pathway. IL-6 induces the activation and unique and powerful approach for the identification of key complex formation of Shc and Grb2 that interacts with Sos1, pathogenetic genes involved in a subset of diseases [78,79]. resulting in Ras activation. Thereafter, the other signaling The observed unique outlier pattern using novel bioinfor- intermediates of the Ras signaling pathway such as Raf-1 and matics approaches is consistent with other translocations MAPK are activated [59]. It has been indicated that the com- where an activating gene can fuse with multiple partners, bination of IL-6 and cellular adhesion of MM cells to fibro- such as the fusion of the immunoglobulin heavy chain pro- nectin via β1 integrins results in a dramatic amplification of moter to CCND1 or FGFR3, in specific subsets of multiple STAT3 phosphorylation, nuclear translocation, and DNA myeloma, respectively [80]. However, meta-analysis supplies binding via a novel preloading of gp130 with STAT3 [76]. a prevailing tool for analyzing microarray experiments by Runt-related transcription factor 2 (Runx2) is a tran- merging data from multiple studies, though it offers unique scription factor that accelerates the formation and differen- computational challenges. Furthermore, several tools with tiation of osteoblasts from mesenchymal stem cells. Myeloma the University of California Santa Cruz (UCSC) genome cells inhibit Runx2 activity and reduce osteoblast differen- browser can be used to screen a locus on the chromosome tiation in osteoprogenitor cells, an effect mediated by both of interest [81-83]. Microarray data analysis using Bioinfor- cell–cell contact and IL-7. In support of a role for Runx2 in matic tools such as DAVID (Database for Annotation, Visu- the pathophysiology of myeloma bone disease, a significant alization and Integrated Discovery) or RankProd may help in reduction in the proportion of Runx2-positive osteoblasts identifying novel genes that are responsible for malignancies, was observed in myeloma patients with evidence of osteolytic such as KRETZSCHMAR_IL6_DIFF and BROCKE_IL6 bone lesions compared to those patients with no evidence of gene sets—gene sets that are differently regulated when MM bone disease [40]. It has been reported that Wnt signaling cells are treated with IL-6 in a model of early infection with promotes osteogenesis by directly stimulating Runx2 gene the protozoan Leishmania chagasi [84]. The Gene Index expression. The inhibitory effect of MM cells on osteoblast and gene-enriched microarray symbolize a precious resource differentiation appears to be mediated by the capacity of for investigators interested in analyzing the molecular basis 34 Abroun et al

FIGURE. 2 Network tools reconstructed from mRNA and miRNA expression data of multiple myeloma samples. FIGURE.3 Bioinformatics tools for promoter analysis can predict genes of functional relevance in cell proliferation in cancer. of MM. Analysis of sequence information with the help of various in silico tools (database of expressed sequence tags [dbEST] or the high-throughput genomic sequence [htgs] are potentially attractive tumor types for MDM2 inhibi- database) revealed numerous poorly characterized genes of tor–based therapy. Shangary and Wang have employed the potential relevance to myeloma biology. It also identified computational 3D screening for the discovery of novel small- the cDNAs necessary to spot a myeloma array. Initial results molecule inhibitors of the MDM2–p53 interaction [92]. using this array demonstrate that it will prove of unique Figure 4 shows the surface representation of the binding value in mining the biology of myeloma [85,86]. The inte- interface of MDM2 complexes with p53. gration of the transcriptional and post-transcriptional levels Computer-aided drug design was introduced in the using artificial network tools reconstructed from mRNA and 1960s, and has quickly developed in 20 years because of its miRNA expression data of MM samples allowed the recog- benefits in the field of life sciences. Many methods, includ- nition of critical genes for both types of regulatory funda- ing the application of various computational programs and mental interactions and analyzing direct relationships at the software, have been developed for the discovery of inhibi- transcriptional level from interaction that are indirect since tors. These approaches can be divided into 2 categories: they are mediated by post- transcriptional regulation (Figures receptor-based design and ligand-based design. In practice, 2 and 3) [87]. mere ligand- or receptor-based studies are often insufficient. Combination approaches quite often are used for improved New Drug Designing performance and outcome. This is true in the development In silico methods and Bioinformatics can support target validation by providing functionally predictive information mined from databases and experimental datasets [88-90]. The prediction of tumor antigens provides potential targets for the immunotherapy of patients with multiple myeloma (MM) and helps in the understanding of carcinogenesis. Zhou et al performed an experiment using Bioinformatic tools to predict 12 novel multiple myeloma special antigen (MMSA) genes, and the antigens identified in their study may be potential candidates for diagnosis of MM and targets for the immunotherapy of this disease [91]. Tumor suppressor p53 is an attractive cancer therapeutic target because it can be functionally activated to eradicate tumors. Because FIGURE.4, Cartoon representation of MDM2 in its apo the activity of MDM2 inhibitors depends upon p53 activa- (shown in red) and holo (shown in green) conformations. tion in cells expressing wild type p53, blood malignancies, B, Surface representation of the binding interface of MDM2 such as MM, which have infrequent p53 mutation/deletion, complexes with p53. Myeloma and Bone Lesion 35

functional genes. Han et al believe that chemokine-like factor superfamily members 1 through 8 represent a novel gene family associated with both chemokine and TM4SF families. Therefore, using in silico and Bioinformatics tools provides new opportunities to investigate the relationship of these 3 families and lays the groundwork for achieving a better understanding of the complicated interactions among them [97].

Survival Signal Pathway Mathematical modeling and simulation techniques have turned out to be valuable tools for the understanding of complex biological systems in different areas of research. By elaborating technical bases for the computer-assisted FIGURE.5, Three-dimensional computer graphic shows a drug modeling in biological systems and experimental techniques candidate (MET tyrosine kinase inhibitor) bound to its tar- such as PyBioS (Max Planck Institute for Molecular Genet- get protein. ics, Munich, Germany), we will be able to study complex pathways such as the Wnt signaling pathway. Wnt signaling is a paracrine interaction, and it acts in numerous cellular of histone deacetylase (HDAC) inhibitors as well. Hideshima processes including cell proliferation, survival, and differen- et al reported that inhibition of HDAC6 causes growth inhi- tiation. It thus has a significant impact on development and bition of MM cells without affecting noncancerous cells [93]. disease [98,99]. Studies have shown that the MET receptor and its ligand hepatocyte growth factor play an important role in the prolif- DISCUSSION eration, migration, adhesion, and survival of MM cells [94]. The in silico method helps to illustrate candidate drug inter- Bone resorption physiologically is done by osteoclasts, action with the MET receptor (Figure 5). and osteoclasts are multi-nuclear cells with a short half-life [100]; considering that MM increases osteoclast activity [45], Cytokines, Chemokines, and Interleukin we cannot apply bone destruction just to osteoclasts because The Bioinformatics and sequence computational stud- osteoclasts form a low percentage of BM cells, and such a ies, using amino acid alphabet propensities and evolutionary vast and continual destruction in bones that are formed dur- trace analyses, suggest several structural regions within the ing several years (40 years) is inconsistent. Thus, we have to fold that are interesting in their amino acid composition or search for other factors that intrigue this destruction, such as in exhibiting class-specific residues. Kanagarajadurai and an increase in specific factors that lead to the differentiation Sowdhamini have done an analysis of the interleukin-8–like of HSCs to osteoclasts, chromosomal translocations that lead chemokine superfamily that provides the first systematic to changes in the activity of some cells, or infections (viral) overview of sequence conservation using objective measures, that lead to abnormal (tumor) plasma cells in MM patients which needs to be studied further. Such sequence-structure– and cause the secretion of inflammatory factors for BM based analyses should be of general value in providing prin- destruction, or any agent that disturbs the balance between ciples and guidelines for dealing with protein super families osteoblasts and osteoclasts. Two microRNAs, 15a and 16, are where the sequence identity is very low [95]. described as a cluster located at chromosome 13q14, which Recognition of cytokines and key cell markers, such as is commonly deleted in more than 50% of MM patients CD4 and CD8, will permit us to initiate specific reagents, [101]. Evaluating involved signaling pathways in excessive or which would then be used to assist us in gaining a better decreasing activity and suppression or accumulation in MM knowledge of the immune response. To overcome the short- patients can help us in medicine production and suitable comings in the automated annotation of the genome and to treatment. For example, although Wnt is the main pathway start to address uncertainties about immune function, Wong for survival of osteoblasts, it has no effect on myeloma cell et al have adopted a manual, expert-curated approach to activity (maybe it has inhibitory effect); increasing serum cal- annotating highly divergent genes using in silico methods. cium and connection to calcium receptors and activation of This was performed using a sensitive TBLASTN search. relevant pathway lead to Wnt inhibition. So we can extrapo- Their success with this strategy suggests that this method will late that the cause of bone destruction, caused by myeloma be applicable to the identification of rapidly evolving gene cells or by increasing in osteoclasts activity, led to the sup- families in other distant vertebrate species [96]. In silico gene pression of BM osteoblasts. In the future it would be interest- identification from the human genome sequence and EST ing to study the calcium effects and activated pathways on database has become critically important in discovering novel BM cells (ie myeloma plasma cells, osteoclasts, osteoblasts, 36 Abroun et al and their progenitors) of MM patients. Maybe a decrease in development: from B-cell subsets to long-term survival niches. vitamin D causes these old individuals to be susceptible to Semin Immunol. 2008;20:49-58. MM. We say that the reason for the accumulation of tumor 15. Sirohi B, Powles R. Multiple myeloma. Lancet. 2004;363:875-887. cells (ie, myeloma cells) in BM may be related to presence of 16. Rahim F. In silico comparison of simple sequence repeats in secreted factors from BM cells accumulating and the pres- high nucleotides-rich genomes of microorganism. Pak J Biol Sci. ence of factors such as iron and calcium, which are detected 2008;11:2372-2381. at high concentrations in BM. Combination therapy will be 17. Chen L, Li J, Xu W, et al. Molecular cytogenetic aberrations in recognized as the most suitable treatment option, especially patients with multiple myeloma studied by interphase fluorescence for patients pursuing poly-therapy to treat co-existing dis- in situ hybridization. Exp Oncol. 2007;29:116-120. ease. Bearing this in mind, we attempted to design in silico 18. Chiecchio L, Dagrada GP, Ibrahim AH, et al. Timing of acqui- systems. The efforts to use in silico systems were successful, sition of deletion 13 in plasma cell dyscrasias is dependent on presenting a feasibility to use such models leading to high genetic context. Haematologica. 2009;94:1708-1713. throughput screening of drugs and their nature. 19. Liebisch P, Döhner H. Cytogenetics and molecular cytogenetics in multiple myeloma. Eur J Cancer. 2006;42:1520-1529. REFERENCES 20. Pratt G. Molecular aspects of multiple myeloma. Mol Pathol. 2002;55:273-283. 1. Raab MS, Podar K, Breitkreutz I, Richardson PG, Anderson KC. 21. Bataille R, Jégo G, Robillard N, et al. The phenotype of normal, Multiple myeloma. Lancet. 2009;374:324-339. reactive and malignant plasma cells. Identification of "many and 2. Criteria for the classification of monoclonal gammopathies, mul- multiple myelomas" and of new targets for myeloma therapy. Hae- tiple myeloma and related disorders: a report of the International matologica. 2006;91:1234-1240. Myeloma Working Group. Br J Haematol. 2003;121:749-757. 22. Brito JL, Walker B, Jenner M, et al. MMSET deregulation affects 3. Kyle RA, Rajkumar SV. Criteria for diagnosis, staging, risk strati- cell cycle progression and adhesion regulons in t(4;14) myeloma fication and response assessment of multiple myeloma. Leukemia. plasma cells. Haematologica. 2009;94:78-86. 2009;23:3-9. 23. Cassinelli G, Ronchetti D, Laccabue D, et al. Concomitant 4. Greipp PR, San Miguel J, Durie BG, et al. International staging downregulation of proliferation/survival pathways dependent on system for multiple myeloma. J Clin Oncol. 2005;23:3412-3420. FGF-R3, JAK2 and BCMA in human multiple myeloma cells by 5. Durie BG, Salmon SE. A clinical staging system for multiple multi-kinase targeting. Biochem Pharmacol. 2009;78:1139-1147. myeloma. Correlation of measured myeloma cell mass with pre- 24. Bisping G, Wenning D, Kropff M, et al. Bortezomib, dexametha- senting clinical features, response to treatment, and survival. Can- sone, and fibroblast growth factor receptor 3-specific tyrosine kinase cer. 1975;36:842-854. inhibitor in t(4;14) myeloma. Clin Cancer Res. 2009;15:520-531. 6. Czaplinski A, Steck AJ. Immune mediated neuropathies—an 25. Reece DE. Management of multiple myeloma: the changing land- update on therapeutic strategies. J Neurol. 2004;251:127-137. scape. Blood Rev. 2007;21:301-314. 7. Blade J. Clinical practice. Monoclonal gammopathy of undeter- 26. Cao S, Liu J, Song L, Ma X. The protooncogene c-Maf is an mined significance.N Engl J Med. 2006;355:2765-2770. essential transcription factor for IL-10 gene expression in macro- 8. Kyle RA, Rajkumar SV. Multiple myeloma. Blood. phages. J Immunol. 2005;174:3484-3492. 2008;111:2962-2972. 27. Athanasiou E, Kaloutsi V, Kotoula V, et al. Cyclin D1 overexpres- 9. Zhan F, Hardin J, Kordsmeier B, et al. Global gene expression pro- sion in multiple myeloma. A morphologic, immunohistochemical, filing of multiple myeloma, monoclonal gammopathy of undeter- and in situ hybridization study of 71 paraffin-embedded bone mined significance, and normal bone marrow plasma cells. Blood. marrow biopsy specimens. Am J Clin Pathol. 2001;116:535-542. 2002;99:1745-1757. 28. Ngo BT, Felthaus J, Hein M, et al. Monitoring bortezomib 10. Magrangeas F, Nasser V, Avet-Loiseau H, et al. Gene expres- therapy in multiple myeloma: screening of cyclin D1, D2, and sion profiling of multiple myeloma reveals molecular por- D3 via reliable real-time polymerase chain reaction and association traits in relation to the pathogenesis of the disease. Blood. with clinico-pathological features and outcome. Leuk Lymphoma. 2003;101:4998-5006. 2010;51:1632-1642. 11. Ocqueteau M, Orfao A, Almeida J, et al. Immunophenotypic 29. Saki N, Abroun S, Farshdousti Hagh M, Asgharei F. Neoplastic characterization of plasma cells from monoclonal gammopathy of bone marrow niche: hematopoietic and mesenchymal stem cells. undetermined significance patients. Implications for the differen- Cell J (Yakhteh). 2011;13:131-6. tial diagnosis between MGUS and multiple myeloma. Am J Pathol. 30. Basak GW, Srivastava AS, Malhotra R, Carrier E. Mul- 1998;152:1655-1665. tiple myeloma bone marrow niche. Curr Pharm Biotechnol. 12. Mårtensson IL, Almqvist N, Grimsholm O, Bernardi AI. The 2009;10:345-346. pre-B cell receptor checkpoint. FEBS Lett. 2010;584:2572-2579. 31. Dib IE, Gressier M, Salle V, Mentaverri R, Brazier M, Kamel 13. Vale AM, Schroeder HW Jr. Clinical consequences of defects in S. Multiple myeloma cells directly stimulate bone resorption in B-cell development. J Allergy Clin Immunol. 2010;125:778-787. vitro by down-regulating mature osteoclast apoptosis. Leuk Res. 14. Fairfax KA, Kallies A, Nutt SL, Tarlinton DM. Plasma cell 2008;32:1279-1287. Myeloma and Bone Lesion 37

32. Sezer O. Myeloma bone disease: recent advances in biology, diag- Blood. 2008;112:374-382. nosis, and treatment. Oncologist. 2009;14:276-283. 53. Vacca A, Ribatti D. Bone marrow angiogenesis in multiple 33. Roodman GD. Targeting the bone microenvironment in multiple myeloma. Leukemia. 2006;20:193-199. myeloma. J Bone Miner Metab. 2010;28:244-250. 54. Mahmoud MS, Huang N, Nobuyoshi M, Lisukov IA, Tanaka 34. Abroun S, Sadeghi B, Poopak B, Hajfathali A, Saki N, Vaghef N. H, Kawano MM. Altered expression of Pax-5 gene in human The role of myeloma cells to osteoclast activation. Cell J (Yakhteh). myeloma cells. Blood. 1996;87:4311-4315. 2010;12:357-62. 55. Ishikawa H, Tsuyama N, Mahmoud MS, et al. CD19 expression 35. Jakob C, Goerke A, Terpos E, et al. Serum levels of total-RANKL and growth inhibition of tumours in human multiple myeloma. in multiple myeloma. Clin Lymphoma Myeloma. 2009;9:430-435. Leuk Lymphoma. 2002;43:613-616. 36. San Miguel JF, Gutiérrez NC, Mateo G, Orfao A. Conven- 56. Shamsasenjan K, Otsuyama K, Abroun S, et al. IL-6-induced activa- tional diagnostics in multiple myeloma. Eur J Cancer. tion of MYC is responsible for the down-regulation of CD33 expres- 2006;42:1510-1519. sion in CD33+ myeloma cells. Int J Hematol. 2009;89:310-318. 37. Terpos E, Dimopoulos MA. Myeloma bone disease: pathophysiol- 57. Haylock DN, Nilsson SK. Osteopontin: a bridge between bone ogy and management. Ann Oncol. 2005;16:1223-1231. and blood. Br J Haematol. 2006;134:467-474. 38. Corre J, Mahtouk K, Attal M, et al. Bone marrow mesenchy- 58. Giuliani N, Colla S, Morandi F, Rizzoli V. Angiopoietin-1 and mal stem cells are abnormal in multiple myeloma. Leukemia. myeloma-induced angiogenesis. Leuk Lymphoma. 2005;46:29-33. 2007;21:1079-1088. 59. Ishikawa H, Mahmoud MS, Fujii R, Abroun S, Kawano MM. 39. Silvestris F, Ciavarella S, De Matteo M, Tucci M, Dammacco F. Proliferation of immature myeloma cells by interleukin-6 is associ- Bone-resorbing cells in multiple myeloma: osteoclasts, myeloma ated with CD45 expression in human multiple myeloma. Leuk cell polykaryons, or both? Oncologist. 2009;14:264-275. Lymphoma. 2000;39:51-55. 40. Edwards CM, Zhuang J, Mundy GR. The pathogenesis of the 60. Ishikawa H, Tsuyama N, Abroun S, et al. Interleukin-6, CD45 bone disease of multiple myeloma. Bone. 2008;42:1007-1013. and the src-kinases in myeloma cell proliferation. Leuk Lymphoma. 41. Saki N, Abroun S, Soleimani M, Kaviani S, Sadeghi B, Vaghef N. 2003;44:1477-1481. The role of myeloma cells in the in vitro differentiation of cord 61. Van Riet I, De Waele M, Remels L, Lacor P, Schots R, Van blood hematopoietic stem cell to osteoclast. Modares J Med Sci Camp B. Expression of cytoadhesion molecules (CD56, CD54, (Pathobiol). 2010;13:53-61. CD18 and CD29) by myeloma plasma cells. Br J Haematol. 42. Aggarwal R, Ghobrial IM, Roodman GD. Chemokines in mul- 1991;79:421-427. tiple myeloma. Exp Hematol. 2006;34:1289-1295. 62. Katz BZ. Adhesion molecules—The lifelines of multiple myeloma 43. Abroun S. Chemokines in homeostasis and cancers. Cell J cells. Semin Cancer Biol. 201;20:186-195. (Yakhteh). 2008;10:155-66. 63. Gazitt Y, Akay C. Mobilization of myeloma cells involves SDF-1/ 44. Abroun S, Ishikawa H, Tsuyama N, et al. Receptor synergy of CXCR4 signaling and downregulation of VLA-4. Stem Cells. interleukin-6 (IL-6) and insulin-like growth factor-I in myeloma 2004;22:65-73. cells that highly express IL-6 receptor alpha [corrected]. Blood. 64. Jakob C, Sterz J, Zavrski I, et al. Angiogenesis in multiple 2004;103:2291-2298. myeloma. Eur J Cancer. 2006;42:1581-1590. 45. Heider U, Fleissner C, Zavrski I, et al. Bone markers in multiple 65. Chen L, Sanderson RD. Heparanase regulates levels of syndecan-1 myeloma. Eur J Cancer. 2006;42:1544-1553. in the nucleus. PLoS One. 2009;4:e4947. 46. Hadjidakis DJ, Androulakis II. Bone remodeling. Ann N Y Acad 66. Al-Quran SZ, Yang L, Magill JM, Braylan RC, Douglas-Nikitin Sci. 2006;1092:385-396. VK. Assessment of bone marrow plasma cell infiltrates in mul- 47. Kawano MM, Ishikawa H, Tsuyama N, et al. Growth mechanism tiple myeloma: the added value of CD138 immunohistochemistry. of human myeloma cells by interleukin-6. Int J Hematol. 2002;76 Hum Pathol. 2007;38:1779-1787. Suppl 1:329-333. 67. Bataille R, Robillard N, Avet-Loiseau H, Harousseau JL, Moreau 48. Sprynski AC, Hose D, Caillot L, et al. The role of IGF-1 as a major P. CD221 (IGF-1R) is aberrantly expressed in multiple myeloma, growth factor for myeloma cell lines and the prognostic relevance of in relation to disease severity. Haematologica. 2005;90:706-707. the expression of its receptor. Blood. 2009;113:4614-4626. 68. Bommert K, Bargou RC, Stühmer T. Signalling and survival path- 49. Menu E, van Valckenborgh E, van Camp B, Vanderkerken K. The ways in multiple myeloma. Eur J Cancer. 2006;42:1574-1580. role of the insulin-like growth factor 1 receptor axis in multiple 69. Krishnan V, Bryant HU, Macdougald OA. Regulation of bone myeloma. Arch Physiol Biochem. 2009;115:49-57. mass by Wnt signaling. J Clin Invest. 2006;116:1202-1209. 50. Nishimoto N. Interleukin-6 as a therapeutic target in candidate 70. Kochegarov A. Small molecules for stem cells. Expert Opin Ther inflammatory diseases.Clin Pharmacol Ther. 2010;87:483-487. Pat. 2009;19:275-281. 51. Gavriatopoulou M, Dimopoulos MA, Christoulas D, et al. Dick- 71. Peterlik M, Grant WB, Cross HS. Calcium, vitamin D and cancer. kopf-1: a suitable target for the management of myeloma bone Anticancer Res. 2009;29:3687-3698. disease. Expert Opin Ther Targets. 2009;13:839-848. 72. Diamond T, Golombick T, Manoharan A. Vitamin D status may 52. Qiang YW, Shaughnessy JD Jr, Yaccoby S. Wnt3a signaling within effect the skeletal complications of multiple myeloma. Am J Hema- bone inhibits multiple myeloma bone disease and tumor growth. tol. 2010;85:302-303. 38 Abroun et al

73. Pearse RN. Wnt antagonism in multiple myeloma: a potential 88. del Val C, Mehrle A, Falkenhahn M, et al. High-throughput pro- cause of uncoupled bone remodeling. Clin Cancer Res. 2006;12(20 tein analysis integrating bioinformatics and experimental assays. Pt 2):6274s-6278s. Nucleic Acids Res. 2004;32:742-748. 74. Komiya Y, Habas R. Wnt signal transduction pathways. Organo- 89. Desany B, Zhang Z. Bioinformatics and cancer target discovery. genesis. 2008;4:68-75. Drug Discov Today. 2004;9:795-802. 75. Desplanques G, Giuliani N, Delsignore R, Rizzoli V, Bataille R, 90. Nagase T, Miyajima N, Tanaka A, et al. Prediction of the coding Barillé-Nion S. Impact of XIAP protein levels on the survival of sequences of unidentified human genes. III. The coding sequences myeloma cells. Haematologica. 2009;94:87-93. of 40 new genes (KIAA0081-KIAA0120) deduced by analysis of 76. Shain KH, Yarde DN, Meads MB, et al. Beta1 integrin adhesion cDNA clones from human cell line KG-1 (supplement). DNA Res. enhances IL-6-mediated STAT3 signaling in myeloma cells: impli- 1995;2:51-59. cations for microenvironment influence on tumor survival and 91. Zhou FL, Zhang WG, Chen G, et al. Serological identification proliferation. Cancer Res. 2009;69:1009-1015. and bioinformatics analysis of immunogenic antigens in multiple 77. Giuliani N, Mangoni M, Rizzoli V. Osteogenic differentiation of myeloma. Cancer Immunol Immunother. 2006;55:910-917. mesenchymal stem cells in multiple myeloma: identification of 92. Shangary S, Wang S. Small-molecule inhibitors of the MDM2-p53 potential therapeutic targets. Exp Hematol. 2009;37:879-886. protein-protein interaction to reactivate p53 function: a novel approach 78. Rubin MA, Chinnaiyan AM. Bioinformatics approach leads to the for cancer therapy. Annu Rev Pharmacol Toxicol. 2009;49:223-241. discovery of the TMPRSS2:ETS gene fusion in prostate cancer. 93. Hideshima T, Bradner JE, Wong J, et al. Small-molecule inhibi- Lab Invest. 2006;86:1099-1102. tion of proteasome and aggresome function induces synergistic 79. Rhodes DR, Yu J, Shanker K, et al. ONCOMINE: a cancer antitumor activity in multiple myeloma. Proc Natl Acad Sci U S A. microarray database and integrated data-mining platform. Neopla- 2005;102:8567-8572. sia. 2004;6:1-6. 94. Phillip CJ, Stellrecht CM, Nimmanapalli R, Gandhi V. Targeting 80. Tomlins SA, Rhodes DR, Perner S, et al. Recurrent fusion of MET transcription as a therapeutic strategy in multiple myeloma. TMPRSS2 and ETS transcription factor genes in prostate cancer. Cancer Chemother Pharmacol. 2009;63:587-597. Science. 2005;310:644-648. 95. Kanagarajadurai K, Sowdhamini R. Sequence and structural 81. Fuellen G. Evolution of gene regulation—on the road towards analyses of interleukin-8-like chemokine superfamily. In Silico Biol. computational inferences. Brief Bioinform. 201;12:122-131. 2008;8:307-330. 82. Flanagan SE, Patch AM, Ellard S. Using SIFT and PolyPhen to 96. Wong ES, Young LJ, Papenfuss AT, Belov K. In silico identifica- predict loss-of-function and gain-of-function mutations. Genet Test tion of opossum cytokine genes suggests the complexity of the Mol Biomarkers. 2010;14:533-537. marsupial immune system rivals that of eutherian mammals. 83. Pruim RJ, Welch RP, Sanna S, et al. LocusZoom: regional visu- Immunome Res. 2006;2:4. alization of genome-wide association scan results. Bioinformatics. 97. Han W, Ding P, Xu M, et al. Identification of eight genes encod- 2010;26:2336-2337. ing chemokine-like factor superfamily members 1-8 (CKLFSF1-8) 84. Ettinger NA, Wilson ME. Macrophage and T-cell gene expression by in silico cloning and experimental validation. Genomics. in a model of early infection with the protozoan Leishmania cha- 2003;81:609-617. gasi. PLoS Negl Trop Dis. 2008;2:e252. 98. Logan CY, Nusse R. The Wnt signaling pathway in development 85. Claudio JO, Masih-Khan E, Tang H, et al. A molecular com- and disease. Annu Rev Cell Dev Biol. 2004;20:781-810. pendium of genes expressed in multiple myeloma. Blood. 99. Moon RT, Kohn AD, De Ferrari GV, Kaykas A. WNT and 2002;100:2175-2186. beta-catenin signalling: diseases and therapies. Nat Rev Genet. 86. Zhou FL, Zhang WG, Chen G, et al. Immunological screening 2004;5:691-701. for multiple myeloma-associated antigens and their bioinformat- 100. Martin TJ. Paracrine regulation of osteoclast formation and activ- ics analysis [in Chinese]. Zhongguo Shi Yan Xue Ye Xue Za Zhi. ity: milestones in discovery. J Musculoskelet Neuronal Interact. 2006;14:252-257. 2004;4:243-253. 87. Biasiolo M, Forcato M, Possamai L, et al. Critical analysis of tran- 101. Roccaro AM, Sacco A, Thompson B, et al. MicroRNAs 15a and scriptional and post-transcriptional regulatory networks in mul- 16 regulate tumor proliferation in multiple myeloma. Blood. tiple myeloma. Pac Symp Biocomput. 2010:397-408. 2009;113:6669-6680.