Negative growth regulators of the cell cycle machinery and cancer

Transcription

1 Pathophysiology 16 (2009) Review Negative growth regulators of the cell cycle machinery and cancer Man Zhang, Huiling Yang Department of Pathophysiology, Sun Yat-Sen University, Guangzhou, PR China Abstract Cell cycle dysregulation is a critical feature of tumor cells. Numerous cell cycle regulators act either as oncogenes or tumor suppressors and their aberrations result in proliferative advantage for cancer cells. Many molecular targets and their use in either abrogating the growth advantage of oncogenic mediators or enhancing the growth suppressive activity of tumor suppressors, have been filed for patents. The molecular targets associated with cell cycle inhibition are of particular interest because they are potential therapeutic agents of promise in the control of inappropriate cellular proliferation. This review focuses on the recent discovery of potential molecular targets involved in cell cycle inhibition and their evaluation as therapeutic agents for cancers. In this review, the strategies employed to control oncogenesis and their possible clinical applications are discussed Elsevier Ireland Ltd. All rights reserved. Keywords: Cancer; Cell cycle; Negative regulator; Signaling transduction Contents 1. Introduction Negative growth regulators Kip family p27 Kip p21 wafl p p Expert opinion References Introduction Cell cycle is defined as a period from the end of one division to the beginning of next division of a proliferative cell. The cell cycle is triggered by an independent cell-cycle control system. At the heart of the control system is a family of protein kinases known as cyclin-dependent kinases (Cdks), Corresponding author at: Department of Pathophysiology, The School of Preclinical Medicine, Sun Yat-Sen University, 74 Zhongshan Road 2, Guangzhou, , PR China. Tel.: ; fax: address: hlyangsums@hotmail.com (H. Yang). whose activity rises and falls as the cell progresses through the cycle. Cdks have no protein kinase activity unless they are tightly bound to a cyclin. In addition, cyclin-cdk complexes can also be regulated by cyclin-dependent kinases inhibitors (CKIs), which bind dramatically and rearrange the structure of the Cdk active site, rendering it inactive. Mammalian cells are known to express several CKIs that contribute to cell-cycle control. According to the difference of structures and functions, CKIs are grouped into two classes: inhibitors of kinase 4 (Ink4) and kinase inhibitory protein (Kip) or Cdk-interacting protein1 (Cip1) or wild-type p53 activated fragment1 (Wafl). Ink4 family bind and inhibit Cdk4/6-cyclin D complexes, and Kip family bind and inhibit /$ see front matter 2009 Elsevier Ireland Ltd. All rights reserved. doi: /j.pathophys

2 306 M. Zhang, H. Yang / Pathophysiology 16 (2009) all Cdk1-, Cdk2-, Cdk4-, and Cdk6-cyclin complexes. The molecular weights of members of Ink4 family are between 15 and 20 kda. Ink4 family include p16 Ink4a, p15 Ink4b, p18 Ink4c and p19 Ink4d. Kip family include three known members: p21 waf1/cip1, p27 Kip1 and p57 Kip2. Our research focused on the Kip family and other negative regulators such as The review mainly introduces the Kip family, and their functions in the cancer. 2. Negative growth regulators 2.1. Kip family Three known members of Kip family include p27 Kip1, p21 waf1/cip1 and p57 Kip2, which are also closely related in sequence. Except different C-terminals, their N-terminals are of high homologies in structure and function and can specifically inhibit several cyclin/cdks activities. Briefly, N- terminal of Kip contains a conservative sequence of 80 amino acids, which bind with cyclin/cdks complex through extracovalent bonds to inhibit Cdk activity. Among these Kips, p27 Kipl and p21 wafl/cip1 are most frequently investigated. p21 wafl/cip1 functions in the response of most mammalian cells to DNA damage. Its synthesis is regulated by p53- dependent and p53-independent pathways p27 Kip1 p27 Kip1 (p27), a CDK inhibitor involved in inhibiting G1 cyclin-cdk activities, causes G1 cell cycle arrest. p27 functions as a new class of tumor suppressor, and is haploinsufficient in tumor suppression. Reduced expression of p27 is frequently detected in many types of human cancers, such as breast, prostate, gastric, lung, skin, colon, and ovarian cancers. Importantly, decreased expression of the p27 protein correlates with cancer development and poor survival. A high percentage of NPC cases have been shown to have a low level of p27 expression by immunohisto-chemistry. In addition, low p27 expression significantly correlates with loco-regional recurrence. In animal models, p27 has been demonstrated to have tumor suppressor activity in lung cancer and Her2-overexpressing engraft cancer models. The activity of p27 Kip1 is regulated in transcription and posttranscriptional level, even at gene level. Although regulating p27 Kip1 may occur on transcriptional level, we found that the mrna level of p27 Kip1 is invariant in most cancer, which suggest that the activity of p27 Kip1 is regulated mainly on post-transcriptional level. Some researches have demonstrated Forkhead [1] may regulate the p27 Kip1 and c-myc [2] inhibits its level. The FOXO family of Forkhead transcription factors, regulated by the phosphoinositide-3-kinase-akt pathway, is involved in cell cycle regulation and apoptosis. FOXO4 was shown to regulate the transcription of the cyclin-dependent kinase inhibitor p27 Kip1 gene directly. Also, we have shown that Her2 promotes mitogenic growth and transformation of cancer cells by down-regulation of p27 Kip1. We hypothesized that an Akt phosphorylation mutant of FOXO4 (FOXO4A3), which maintained the activity to transactivate p27 Kip1, may be used as an anticancer agent for Her2-overexpressing cancers. We found that FOXO4A3 inhibited the kinase activity of protein kinase B/Akt and reversed Her2-mediated p27 Kip1 mislocation in the cytoplasm. FOXO4A3 expression also led to decreased levels of CSN5, a protein involved in p27 Kip1 degradation. These data suggested that FOXO4A3 also can regulate p27 Kip1 post-transcriptionally. In addition, we found that FOXO4A3 sensitized cells to apoptosis induced by the chemotherapeutic agent 2-metho-xyestradiol (2ME). FOXO4A3 expression in Her2-overexpressing cells can be regulated in vivo and reduces the tumor volume in a tumor model. These findings indicate the applicability of employing FOXO4 regulation as a therapeutic intervention in Her2-overexpressing cancers [3]. Our study has shown that Her2 signaling can enhance p27 Kip1 ubiquitination, thereby promoting p27 Kip1 degradation and subsequent activation of Cdk activity, and degradated p27 Kip1 by ubiquitination be enhanced by JAB1 binding as well as by phosphorylation on Thr187. So we generated modified p27 Kip1 proteins, which were mutated at Thr187 or deleted at JAB1 binding domain, induction of p27 T187A and p27 T187A DJAB inhibits Her2-activated cell growth, Cdk2 activity, cell proliferation and transformation. Significantly, a modified protein (p27 T187A DJAB) reduced the tumor volume in a Her2-overexpressing tumor model efficiently. These findings demonstrate modified p27 Kip1 proteins may be used as a therapeutic intervention in Her2-overexpressing cancers [4]. Besides, Akt is a crucial regulator of oncogenic signal and can phosphorylate p27 Kip1 to enhance p27 Kip1 degradation, thereby promoting cell growth. Our research has shown that mediated cell cycle arrest concurred with p27 Kip1 up-regulation and Akt inactivation. We found that blocks Akt-mediated acceleration of p27 Kip1 turnover rate inhibited Akt-mediated p27 Kip1 phosphorylation that targeted p27 Kip1 for nuclear export and degradation. Low expression of in human primary breast cancers correlated with cytoplasmic location of p27 Kip1. These data provide an insight into activity and rational cancer gene therapy by identifying as a positive regulator of p27 Kip1 and as a potential anticancer agent [5]. We also found that alternative reading frame protein (ARF), a tumor suppressor protein encoded by a gene located in the Ink4a/ARF gene locus, was frequently inactivated in human cancers. We applied the ARF gene as a tumorsuppressive agent for Her2/neu-overexpressing cells under the control of a tetracycline (tet)-regulated gene expression system. We found that ARF antagonized Akt-mediated p27 Kip1 phosphorylation and increased p27 Kip1 stability in Her2/neu-overexpressing cells. ARF expression also led to decreased levels of Cul1 and Skp2, two proteins involved in p27 Kip1 degradation. We also found that ARF caused apoptosis in Her2/neu-overexpressing cells, and sensitized cells to

3 M. Zhang, H. Yang / Pathophysiology 16 (2009) apoptosis induced by the chemotherapeutic agents paclitaxel and 2-Me [6] p21 wafl p21 wafl/cip1 responds to DNA damage of most mammalian cells. Its synthesis is regulated by p53-dependent and p53- independent pathways. p21 waf1, a down-stream target of p53 gene, presents a dual inhibitory effect on cell cycle, that is, the binding of its N-terminal with Cdks to prevent cells from entering S phase and that of its C-terminal with PCNA to block DNA replication. The reduced p21 waf 1 mrna expression is frequently found in tumor cells, but the mutation of p21 waf 1 in human cancer is very rare like other CKIs. We found that manumycin (a far-nesyltransferase inhibitor) induced endogenous expression of p21 waf1 in anaplastic thyroid cancer cells. Manumycin increased the activity of the p21 waf1 promoter, the level of p21 waf1 mrna and the amount of p21 waf1 protein. We hypothesized that p21 waf1 had a proapoptotic effect in cells treated with manumycin, or paclitaxel, or both agents. By measuring viability and caspase-3 activity, we found that stably transfected KAT-4 cells with p21 waf1 cdna under the control of a metallothionein promoter were more sensitive to manumycin alone, paclitaxel alone, and manumycin plus paclitaxel when p21 waf1 was induced. The increased sensitivity of the cells with induced p21 was associated with an increase in caspase- 3 activity (i.e. apoptosis). We also found that cells with both p21 waf1 alleles deleted were less sensitive to manumycin plus paclitaxel than its wild-type parent cells. Expression of p21 waf1 did not induce apoptosis but enhanced the cytotoxic effects of manumycin and paclitaxel. These findings suggested that p21 waf1 might be required to maintain cell sensitivity to the cytotoxic effects of manumycin and paclitaxel [7]. In recent years, antisense oligonucleotides have been found to be effective tools to inhibit the expression of specific gene. The p21 waf1 antisense oligodeoxynucleotides (p21 AS ODNs) and the random control oligodeoxynucleotides (p21 RD ODNs) were synthesized. Then we successfully transfected p21 AS ODNs and p21 RD ODNs into CNE-1- wt p53 nasopharyngeal carcinoma cell line. The manipulations included: (1) The protein expression levels of p21 waf1 were evaluated using Western blotting analysis. (2) Cell cycle progression and apoptotic cells were assessed by flow cytometric analysis. (3) The clonogenic survival assay was performed to determine the survival fraction. (4) The parameters D0, Dq, and N for the single-hit multitarget model and the parameters a, b, a/b, and SF2 for the linear quadratic model were calculated. BALB/c nude mice were used to investigate the effect of p21 waf1 AS ODNs on the radiosensitivity of nasopharyngeal xenografts in vivo. Finally, we demonstrated our hypothesis that p21 waf1 Antisense oligodeoxynucleotides led to inhibition of p21 waf1 protein expression, loss of G1 cell cycle arrest, increase of apoptosis in CNE-1- wt p53 nasopharyngeal carcinoma cell line in vitro and inhibited tumor growth in vivo. Antisense oligodeoxynucleotides may become a promising strategy to enhance radiosensitivity in nasopharyngeal carcinoma cells with normal p53 function [8] p53 p53 is a transcription factor that establishes programmers for apoptosis, and repair in response to a variety of cellular stresses, including DNA damage, hypoxia and nutrient deprivation [9]. p53 include mutation and wild-type. Wildtype p53 ( wt p53) is critically important for operation of the DNA damage-induced checkpoint. On sensing DNA damage, wt p53 is activated, resulting in either G1 cell cycle arrest or apoptosis, which allow time for the cell to either repair the damage or rid the body of cells with damaged DNA. Loss of p53 function, therefore, decreases genomic stability, and thus increase the accumulation of additional genetic mutations required for neoplastic transformation. Germ line p53 mutation is involved in the cancer-prone Li-Fraumeni syndrome The p53 gene is the most frequently mutated gene in human cancers. By the way, the Mdm2 (murine double minute oncogene) gene encodes a protein that binds p53 and targets it destruction by the ubiquitin proteosome pathway. Too much Mdm2 protein may be analogous to p53 inactivation because any p53 synthesized would be rapidly degraded. Overexpression of Mdm2 mediated by gene amplification can also be detected in human cancer, particularly sarcoma. Our results have shown that CSN5, the 5th subunit of COP9 signalosome, be a critical regulator of both p53 and Mdm2. Significantly, we further show that CSN5 antagonize the transcriptional activity of p53 and curcumin, an important inhibitor of CSN-associated kinases, can down-regulate not only CSN5 but also Mdm2, which resulted in p53 stabilization [10]. Recently, we found that was a critical regulator of Mdm can affect the location of Mdm2 from the nucleus to the cytoplasm, and that Mdm2-mediated cytoplasmic localization of p53 can be reversed by the presence of , Mdm2 down-regulation, which stabilized p53 and inhibited tumor growth in animal tumors, blocked Mdm2-mediated Rb degradation and p53 NEDDylation [11]. Our studies may pave the way of stabling p53 for anticancer drug p57 p57, which is related to p21 Cip1 and p27 Kip1, is a potent, tight-binding inhibitor of several G1 cyclin/cdk complexes, and its binding is cyclin dependent. Unlike Cip1, Kip2 is not regulated by p53. Overexpression of p57 Kip2 arrests cells in G1. p57 Kip2 proteins have a complex structure [12]. Mouse p57 Kip2 consists of four structurally distinct domains: an amino-terminal Cdk inhibitory domain, a proline-rich domain, an acidic-repeat region, and a carboxy-terminal domain conserved with p27 Kip1. Human p57 Kip2 appears to have conserved the amino- and carboxy-terminal domains but has replaced the internal regions with sequences containing proline alanine repeats. In situ hybridization during mouse embryogenesis revealed that Kip2 mrna displays a striking pattern of expression during development, showing

4 308 M. Zhang, H. Yang / Pathophysiology 16 (2009) high-level expression in skeletal muscle, brain, heart, lungs, and eye. Most of the Kip2-expressing cells are terminally differentiated, suggesting that p57 Kip2 is involved in decisions to exit the cell cycle during development and differentiation. Human Kip2 is located at 11p15.5, a region implicated both in sporadic cancers and Beckwith Wiedemann syndrome, a familial cancer syndrome, marking it as a candidate tumor suppressor. The discovery of a new member of the p21 Cip1 inhibitor family with novel structural features and expression patterns suggests a complex role for these proteins in cell-cycle control and development σ The protein, a negative regulator of the cell cycle [13,14], is a human mammary epithelium-specific marker that is down-regulated in transformed mammary carcinoma cells [15]. It has also been identified as a p53-inducible gene product involved in cell cycle checkpoint control after DNA damage [16] and stabilizes the expression of p53. We found that there was a failure to up-regulate in response to DNA damage in two NPC cell lines that have p53 mutation. Its low expression and function in some certain tumor plays an important role in tumorigenesis, including breast cancer [17], gastric cancer [18], hepatocellular carcinoma [19], lung cancer [20] and nasopharyngeal carcinoma [21]. Our evidence was provided that binds and inhibits Akt. Low expression of in human primary breast cancers correlated with Akt activation. Significantly, we have shown that inhibit Akt-mediated cell growth, transformation and tumorigenesis. Previous studies have shown that it can negatively regulate cell cycle progression by inhibiting the activities of Cdk2 [22,23]. We showed that was a critical regulator of Mdm2 at the RING domain. The C-terminal region of binded to Mdm2 very efficiently. Importantly, overexpression led to destabilization of Mdm2 through enhancing Mdm2 self-ubiquitination and accelerating turnover rate. Conversely, loss of resulted in a significant increase in Mdm2 protein. Moreover, live-cell images indicated that can affect the location of Mdm2 from the nucleus to the cytoplasm, and that Mdm2-mediated cytoplasmic localization of p53 can be reversed by the presence of Significantly, we further show that caused Mdm2 down-regulation, thereby stabilizing p53 and inhibiting tumor growth in animal tumors. Also, blocked Mdm2-mediated retinoblastoma degradation and p53 NED- Dylation. We provided evidence that was a pivotal Mdm2 regulator involved in blocking a variety of activities of Mdm2, and affect the tumorigenesis [11]. In addition, we found that sensitized NPC cells to apoptosis induced by the chemotherapeutic agent 2-methoxyestradiol. Further, our finding that sensitized NPC cells to chemotherapeutic drug-mediated apoptosis was reminiscent of the observation that Akt mouse embryonic fibroblasts were more susceptible to apoptosis stimuli than wild-type Akt mouse embryonic fibroblast cells. These findings provided an insight into the roles of and suggested that approaches that modulate activity may be useful in the treatment of NPC and other cancers [21]. These data also provide an insight into rational cancer gene therapy by identifying as a molecular regulator of Akt and as a potential anticancer agent for Akt-activated cancers [24]. 3. Expert opinion Cancer has long been considered to be an endlessly adaptable and profoundly complex disease treatable only with blunt approaches that frequently do as much damage to the patient as to the tumor. Cancer arise when the molecular network connecting proliferation and tumor suppression become uncoupled. Even then, however, the underlying tumor-suppressor programmes remain intact, awaiting only adroit human intervention to reconnect them and herald a new era of effective and tumor-specific therapies. In this review, we have discussed many modes of inhibiting the cell cycle for the potential treatment of cancer. It is important to point out that these rational cancer therapeutic approaches have been prosperously developed within the past 3 years. Some molecular targets that were patented earlier for antitumor therapeutics are equally important and are already progressing in terms of clinical trials, although they are not discussed here. For example, the most important tumor suppressor, p53, has been successfully employed as a molecular target for cancer gene therapy [11,25]. p53 gene therapy for cancer is currently undergoing phase I/II clinical trials in several countries [26]. p53 has been shown to have therapeutic efficacy against a wide range of human tumors containing non-functional p53, thus offering a novel strategy for suppressing tumor growth [27,28]. These inhibitors can also inhibit cell cycle progression by inhibiting the activities of their targets. For example, some of these proteasome inhibitors are in the process of clinical trials [29] and cause accumulation of CKIs such as p21 waf1 and p27 Kip1 to inhibit tumor cell growth [30], and even to help cells overcome drug resistance [31]. The ultimate goal of researchers of the cell cycle and cancer is to study cancer formation in sufficient molecular detail to design rational intervention and cure the disease. With the newly increased mechanistic insights into the cancer cell cycle, the potential for future application of the discussed approaches in cancer therapy is optimistic. We know these approaches could change the treatment of cancer following more commitment and effort from the biotechnology industry and clinical trials. References [1] P.F. Dijkers, R.H. Medema, C. Pals, et al., Forkhead transcription factor FKHRL1 modulates cytokine-dependent transcriptional regulation of p27 [J], Mol. Cell. Biol. 20 (24) (2000)

5 M. Zhang, H. Yang / Pathophysiology 16 (2009) [2] W. Yang, J. Shen, M. Wu, et al., Repression of transcription of the p27 cycle-dependent kinase inhibitor gene by c-myc [J], Oncogene 20 (14) (2001) [3] H. Yang, R. Zhao, H.-Y. Yang, et al., Constitutively active FOXO4 inhibits Akt activity, regulates p27 Kip1 stability, and suppresses HER2- mediated tumorigenicity [J], Oncogene 24 (2005) [4] H.-Y. Yang, H. Yang, R. Zhao, et al. Modified p27 Kip1 is efficient in suppressing HER2-mediated tumorigenicity [J]. J. Cell. Biochem. 98, [5] H. Yang, Y. Zhang, R. Zhao, et al., Negative cell cycle regulator stabilizes p27 Kip1 by inhibiting the activity of PKB/Akt [J], Oncogene (2006) [6] Y. Zhang, X.-C. Zhang, H. Yang, et al., Tumor suppressor ARF inhibits HER-2/neu-mediated oncogenic growth [J], Oncogene 23 (2004) [7] H.-L. Yang, J.-X. Pan, L. Sun, et al., p21 Waf 1 (Cip-1) enhances apoptosis induced by manumycin and paclitaxel in anaplastic thyroid cancer cells [J], J. Clin. Endocrinol. Metab. 88 (2) (2005) [8] X.-F. Liu, Y.-F. Xia, H.-L. Yang, et al., The effect of p21 antisense oligodeoxynucleotides on the radiosensitivity of nasopharyngeal carcinoma cells with normal p53 function [J], Cell Biol. Int. 30 (2006) [9] S.W. Lowe, E. Cepero, G. Evan, Intrinsic tumour suppression [J], Nature 18 (432) (2004) [10] X.-C. Zhang, J. Chen, C.-H. Su, H.-Y. Yang, M.-H. Lee, Roles for CSN5 in control of p53 Mdm2 activities [J], J. Cell. Biochem. 103 (4) (2008) [11] H.-Y. Yang, Y.-Y. Wen, Y.-L. Lin, L. Pham, C.-H. Su, H. Yang, J. Chen, M.-H. Lee, Roles for negative cell regulator in control of MDM2 activities [J], Oncogene 26 (52) (2007) [12] S. Matsuoka, M.C. Edwards, C. Bai, et al., p57 Kip2, a structurally distinct member of the p21 Cip1 Cdk inhibitor family, is a candidate tumor suppressor gene [J], Genes Dev. 9 (6) (1995) [13] C. Laronga, H.Y. Yang, C. Neal, M.H. Lee, Association of the cyclin-dependent kinases and negatively regulates cell cycle progression [J], J. Biol. Chem. 275 (2000) [14] T.A. Chan, H. Hermeking, C. Lengauer, et al., is required to prevent mitotic catastrophe after DNA damage [J], Nature 401 (1999) [15] M.H. Lee, H.Y. Yang, Negative regulators of cyclin dependent kinases and their roles in cancers [J], Cell. Mol. Life Sci. 58 (2001) [16] H.-Y. Yang, Y.-Y. Wen, C.-H. Chen, et al., positively regulates p53 and suppresses tumor growth [J], Mol. Cell. Biol. 23 (20) (2003) [22] C. Laronga, H.Y. Yang, C. Neal, et al., Association of the cyclindependent kinases and negatively regulates cell cycle progression [J], Biology 275 (2000) [23] M.-H. Lee, H.-Y. Yang, Regulators of G1 cyclin-dependent kinases and cancers [J], Cancer Metastasis Rev. 22 (2003) [17] A.T. Ferguson, E. Evron, C.B. Umbricht, et al., High frequency of hypermethylation at the locus leads to gene silencing in breast cancer [J], Proc. Natl. Acad. Sci. U.S.A. 97 (2000) [18] H. Suzuki, F. Itoh, M. Toyota, et al., Inactivation of the gene is associated with 5 CpG island hypermethylation human cancer [J], Cancer Res. 60 (2000) [19] N. Iwata, H. Yamamoto, S. Sasaki, et al., Frequent hypermethylation of CpG islands and loss of expression of the gene in human hepatocellular carcinoma [J], Oncogene 19 (2000) [20] H. Osada, Y. Tatematsu, Y. Yatabe, et al., Frequent and histological typespecial inactivation of in human lung cancers [J], Oncogene 21 (2002) [21] H. Yang, R. Zhao, M.-H. Lee, , a p53 regulator, suppresses tumor growth of nasopharyngeal carcinoma [J], Mol. Cancer Ther. 5 (2) (2006) [24] H. Yang, Y.-Y. Wen, R. Zhao, et al., DNA damage-induced protein inhibits protein kinase B/Akt activation and suppresses Aktactivated cancer [J], Cancer Res. 66 (6) (2006) [25] P. Chene, p53 as a drug target in cancer therapy [J], Expert Opin. Ther. Patents 11 (2001) [26] S.G. Swisher, J.A. Roth, p53 gene therapy for lung cancer [J], Curr. Oncol. Rep. 4 (2002) [27] L.L. Nielsen, D.C. Maneval, p53 tumor suppressor gene therapy for cancer [J], Cancer Gene Ther. 5 (1998) [28] J.A. Roth, S.G. Swisher, R.E. Meyn, p53 tumor suppressor gene therapy for cancer [J], Oncology 13 (1999) [29] J. Adams, Preclinical and clinical evaluation of proteasome inhibitor PS-341 for the treatment of cancer [J], Curr. Opin. Chem. Biol. 6 (2002) [30] J. Sun, S. Nam, C.S. Lee, B. Li, et al., CEP1612, a dipeptidyl proteasome inhibitor, induces p21 waf1 and p27 Kip1 expression and apoptosis and inhibits the growth of the human lung adenocarcinoma A-549 in nude mice [J], Cancer Res. 61 (2001) [31] T. Hideshima, P. Richardson, D. Chauhan, et al., The proteasome inhibitor PS-341 inhibits growth, induces apoptosis, and overcomes drug resistance in human multiple myeloma [J], Cancer Res. 61 (7) (2001)