Clinical and biological significance of microRNA profiling in patients with myeloma

8539

MicroRNA, a small endogenous RNA regulating specific expressed gene function has been implicated in normal biological processes as well as in malignant transformation. Here we have investigated the role of microRNAs in multiple myeloma (MM) biology, and their influence on prognosis and clinical outcome. We evaluated profiles of 384 microRNAs in CD138+ MM cells from 79 patients with MM, 11 cell lines and 9 healthy donors using qRT-PCR based microRNA array. We detected significant modulation of expression of 61 microRNAs in myeloma cells compared to normal plasma cells. Unsupervised hierarchical clustering of filtered matured microRNAs, identified two major groups within the MM population (groups A and group B). Group A clustered with MM cell lines, indicating more aggressive course of disease. Within B group, second degree node, group B2, clustered with normal plasma cells indicating indolent course. The unsupervised clustering of all MM samples showed consistent change in miR-146b, -140, -145, -125a, -151, -223, -155, let-7f, indicative of a role of these microRNA in myelomagenesis; while supervised analysis of samples within groups A and B identified modulated expression of different sets of miRNAs. In group A miR585 and let-7f were upregulated 8–12 fold; in group B, all differentially expressed microRNAs were downregulated (p<0.001) compared to normal plasma cells. These modulated miRNAs target critical signaling molecules such as HOX9, c-myc, Bcl-2, SHP1 and SHP2. We further analyzed the effect of microRNA on clinical outcome. We have observed significantly superior event free and overall survival of patients in group B compared to patients ingroup A (2 yr estimated EFS 79% versus 54% respectively; p=0.05; and 2 yr estimated OS 94% versus 70% respectively; p =0.017). Functional analysis by modulating miRNAs 585, 155 and let-7f showed change in levels of predicted genes with consequent biological effect on growth and apoptosis in MM cell line. Taken together, this data identifies miRNAs as critical modulators of gene expression and signaling pathways and provides potential novel targets in MM to both understand biological behavior and for therapeutic application.

No significant financial relationships to disclose.

TNF-a, retinoid acid and STAT4 pathways are differentially regulated by the HDAC inhibitors, SAHA, TSA and Sirtinol in Waldenstrom's Macroglobulinemia

e14582

Background: HDAC inhibitors (HDAC-I) have emerged as a promising class of therapeutics for the treatment of lymphoplasmacytic malignancies, particularly in combination strategies. The basic mechanisms by which these agents modulate malignant cell growth remain to be clarified. In recent studies, we showed that SAHA, TSA, and Sirtinol have significant preclinical activity in Waldenstrom's macroglobulinemia (WM). As part of our effort to identify potential molecular targets for these agents, we performed gene expression profiling (GEP) using WM cells treated with these agents. Methods: We implemented the Bootstrap re-sampling technique on five BCWM1 GEP datasets to characterize the range of gene expression observed. Individual BCWM1 cells were treated with SAHA, TSA and Sirtinol at their established IC50 dosages for 6 and 24 hours prior to gene expression profiling. The genes which changed by two-fold, and fell outside of the 90% range of expression after drug treatment were identified. Results: SAHA upregulated TNF-α and downstream proteins including HSP70, TNFSF9, SERPINB2, as well as STAT4- dependent genes. Sirtinol upregulated metalloproteinase (MMP) and phosphodiesterase (PDE) family members through retinoic acid signaling. We also observed that SAHA, TSA and Sirtinol differentially upregulate elements of the MAP kinase pathways, including signaling through Erk, JNK, and p38 MAPK, indicating a global stress response to HDAC inhibition. Conclusions: The above studies suggest that HDAC inhibitors differentially effect growth of WM cells through both global as well as pathway specific alterations in signaling. Importantly, this data provides a platform for studying the mechanistic role of HDACs in WM, and provide a framework for the use of HDAC-I as single and combination.

No significant financial relationships to disclose.

IgG and IgA hypogammaglobulinemia is pervasive in Waldenstrom's macroglobulinemia (WM) and persists regardless of response or type of therapeutic intervention

8042

Background: IgG and IgA hypogammaglobulinemia (HG) is a commonly observed in patients with WM and has often been attributed to the extensive infiltration of lymphoplasmacytic cells in the bone marrow (BM). Methods: In this retrospective study we analyzed immunoglobulin and BM infiltration levels from 196 previously untreated patients at their first visit to our clinic who met the clinical and pathological criteria for WM. In addition, we also serially evaluated 65 WM patients who received various therapies for changes in IgA and IgG. All lab values and biopsies used were conducted at our clinic at the Dana-Farber Cancer Institute to insure consistency. Pre-therapy, end of therapy, and best Ig response post therapy were evaluated and patients were followed up to the point of their next treatment regimen. Results: Of the 197 previously untreated patients, 117 (59.7%) demonstrated abnormally low levels of IgG (<700 mg/dl), whereas 131 (66.8%) had abnormally low levels of IgA (<70 mg/dl), and 102 (52.0%) had abnormally low levels of both IgG and IgA. Median value for IgM was 2820 mg/dl (Range 179 - 12400), for IgG 623 mg/dl (Range 127–2770), for IgA 48.5 mg.dl (range 7 - 509), and for the 94 patients for whom BM biopsies were available, the median intratrabecular space involvement was 37.5% (range 5 - 95). No correlation was found between BM involvement and any Ig level. Results from each of the trials is summarized in the table below. Median follow up was 455 days. No correlation between degree of response and Ig recovery was observed. No Ig level that started abnormally low was ever restored to normal levels in this study despite the fact that some recovery is noted in the non-rituximab based trials. Conclusions: These studies indicate that there is no relationship between BM involvement HG in WM. Moreover, therapeutic intervention does not restore IgG or IgA levels indicating a possible constitutive Ig dysregulation problem in patients with WM.

No significant financial relationships to disclose.

[Table: see text]

SHP2 mediates the protective effect of interleukin-6 against dexamethasone-induced apoptosis in multiple myeloma cells

Our previous studies have shown that activation of a related adhesion focal tyrosine kinase (RAFTK) (also known as Pyk2) is required for dexamethasone (Dex)-induced apoptosis in multiple myeloma (MM) cells and that human interleukin-6 (IL-6), a known growth and survival factor for MM cells, blocks both RAFTK activation and apoptosis induced by Dex. However, the mechanism whereby IL-6 inhibits Dex-induced apoptosis is undefined. In this study, we demonstrate that protein-tyrosine phosphatase SHP2 mediates this protective effect. We show that IL-6 triggers selective activation of SHP2 and its association with RAFTK in Dex-treated MM cells. SHP2 interacts with RAFTK through a region other than its Src homology 2 domains. We demonstrate that RAFTK is a direct substrate of SHP2 both in vitro and in vivo, and that Tyr(906) in the C-terminal domain of RAFTK mediates its interaction with SHP2. Moreover, overexpression of dominant negative SHP2 blocked the protective effect of IL-6 against Dex-induced apoptosis. These findings demonstrate that SHP2 mediates the anti-apoptotic effect of IL-6 and suggest SHP2 as a novel therapeutic target in MM.

RAFTK/PYK2-dependent and -independent apoptosis in multiple myeloma cells

Related Adhesion Focal Tyrosine Kinase (RAFTK; also known as Pyk2), is a member of the Focal Adhesion Kinase (FAK) subfamily and is activated by TNF alpha, UV light and increases in intracellular calcium levels. However, the function of RAFTK remains largely unknown. Our previous studies demonstrated that treatment with dexamethasone (Dex), ionizing radiation (IR), and anti-Fas mAb induces apoptosis in multiple myeloma (MM) cells. In the present study, we examined the potential role of RAFTK during induction of apoptosis in human MM cells triggered by these three stimuli. Dex-induced apoptosis, in contrast to apoptosis triggered by anti-Fas mAb or IR, is associated with activation of RAFTK. Transient overexpression of RAFTK wild type (RAFTK WT) induces apoptosis, whereas transient overexpression of Kinase inactive RAFTK (RAFTK K-M) blocks Dex-induced apoptosis. In contrast, transient overexpression of RAFTK K-M has no effect on apoptosis triggered by IR or Fas. In Dex-resistant cells, Dex does not trigger either RAFTK activation or apoptosis. Finally, interleukin-6 (IL-6), a known survival factor for MM cells, inhibits both activation of RAFTK and apoptosis of MM.1S cells triggered by Dex. Our studies therefore demonstrate Dex-induced RAFTK-dependent, and IR or Fas induced RAFTK-independent apoptotic signaling cascades in MM cells.

Functional interaction between retinoblastoma protein and stress-activated protein kinase in multiple myeloma cells

Previous studies have demonstrated that gamma-irradiation (IR)-induced apoptosis in multiple myeloma (MM) is associated with activation of stress-activated protein kinase (SAPK). In the present study, we examined the molecules downstream of SAPK/C-Jun N-terminal kinase (JNK), focusing on the role of retinoblastoma protein (Rb) during IR-induced MM cell apoptosis. The results demonstrate that IR activates SAPK/JNK, which associates with Rb both in vivo and in vitro. Far Western blot analysis confirms that SAPK/JNK binds directly to Rb. IR-activated SAPK/JNK phosphorylates Rb, and deletion of the phosphorylation site in the COOH terminus domain of Rb abrogates phosphorylation of Rb by SAPK/JNK. Taken together, our results suggest that Rb is a target protein of SAPK/JNK and that the association of SAPK/JNK and Rb mediates IR-induced apoptosis in MM cells.

Dexamethasone induces apoptosis of multiple myeloma cells in a JNK/SAP kinase independent mechanism

The stress-activated protein kinases (SAPKs), also known as c-Jun amino-terminal kinases (JNKs), are activated in response to diverse stimuli including DNA damage, heat shock, interleukin-1, tumor necrosis factor-alpha and Fas. Although all these inducers cause apoptosis, whether SAPK/JNK activation is required for apoptosis is controversial. In this study, we demonstrate that ionizing radiation (IR) and dexamethasone (Dex) induce apoptosis in multiple myeloma (MM) derived cell lines, as well as in patient cells. IR-induced apoptosis is associated with activation of SAPK/JNK and p38 kinase, in contrast to Dex-induced apoptosis, which is not associated with activation of stress kinases. Moreover, Dex-induced apoptosis is associated with a significant decrease in the activities of mitogen activated protein kinase (MAPK) and p70S6K, whereas IR-treatment does not alter the activity of these kinases. Both IR and Dex induce poly (ADP ribose) polymerase (PARP) cleavage, a signature event of apoptosis. Finally, interleukin-6 (IL-6) inhibits Dex-induced apoptosis, downregulation of MAP and p70S6K growth kinases and PARP cleavage; in contrast, IL-6 does not inhibit IR-induced apoptosis, activation of SAPK/JNK, and PARP cleavage. Taken together, our findings suggest that SAPK/JNK activation is not required for apoptosis in MM cells, and that there are at least two distinct apoptotic signaling pathways: (i) SAPK/JNK-associated, which is induced by IR and unaffected by IL-6; and (ii) SAPK/JNK-independent, which is induced by Dex, associated with downregulation of MAPK and p70S6K and inhibited by IL-6.