Effects Of Deleted Mutant IκBα On The Apoptosis And Differentiation Of HL-60 Leukemic Cells And Its Mechanism

Nuclear factor-κB (NF-κB) is a pivotal nuclear transcription factor,whose abnormal elevations of activity occur in many tumors. Onceactivation, NF-κB not only governs the expression of genes involved inthe cell growth and proliferation. More important, it also upregulatessome anti-apoptosis genes expression, leading to the tumor cells to escapefrom apoptosis. Therefore, NF-κB is thought as one of the key factors ofsurvival for tumor cells. Since a variety of human leukemic cellsincluding primary acute myeloid leukemia (AML) blast cells, leukemicstem cells (LSCs) and cell lines such as HL-60 cell show constitutiveactive NF-κB, NF-κB is suggested to take part in the leukemogenesis, andNF-κB modulation thus may be a plausible therapeutic target for AML.NF-κB activation is controlled by many factors, of which the IκBαisone of the main inhibitor of NF-κB activation. Generally, NF-κB issequestered in the cytoplasm in an inactive form through its associationwith IκBα. Upon cellular stimulation with diverse agents, IκBαis rapidlyphosphorylated by IκB kinase (IKK) complex, Phosphorylation of IκBαprotein is then followed by ubiquitination—the covalent attachment ofmultiple ubiquitin molecules, and subsequently IκBαis degraded byproteasome, allowing the release of NF-κB, thus the nuclear localizationsequence of NF-κB is unmasked. This process triggers translocation ofNF-κB from the cytoplasm to the nucleus where it stimulatestranscription of target genes withκB binding site. Obviously, thephosphorylation and degradation of IκBαare the two key stepsresponsible for the regulation of NF-κB activation, inhibition of IκBαphosphorylation and degradation might block the activation of NF-κB,and affect its biologic functions, which has been evidenced by a great ofstudies. For example, NF-κB activation in many solid tumor andleukemic cells was inhibited by a variety of mutant IκBα(mIκBα), and apoptosis was simultaneously induced. In addition, inhibitors ofproteasome such as MG-132, can inhibit the NF-κB activation in LSCssorted from the patients with AML and induce LSCs apoptosis. However,like other inhibitors of proteasome, MG-132 is also not a direct specificinhibitor against NF-κB activation. Furthermore, high concentration orlong incubation (＞24 hours), normal CD34⁺ cells are eventually impairedby treatment with MG-132. Thus, the using of proteasome inhibitors toinhibit NF-κB activation may be not the best strategy. Although, therewas reported that NF-κB activation was inhibited by mIκBα, and in oneor two of these studies, it was generally to use conventional Ad5-basedadenovirus (Ad)-mediated expression of a modified nondegraded form ofIκBαin which Ser32 and Ser36 are substituted with nonphosphorylatablealanine. However, some studies have demonstrated that amino acids atother sites of IκBαcan also be phosphorylated in addition to thephosphorylations at Ser32 and Ser36 of IκBα, one case showed thattyrosine 42 of IκBαcan be phosphorylated by nonreceptor tyrosinekinases. What is more important is that transferring of gene into earlyhematopoietic cells such as hematopoietic stem/progenitor cells andleukemia blast cells by Ad5-based Ad vectors (AdVec) infection has beengenerally problematic. Indeed, a number of investigators have noted thatbone marrow cells and some malignant myeloid cell lines were refratoryto Ad5-based AdVec infection. To overcome the disadvantages, lookingfor some alternative methods is required.In view of current works, in this study, the deleted mIκBαcDNAwas transferred into HL-60 cells by a novel chimeric Ad5F35 AdVecinfection to investigate the effects of the mIκBαon the apoptosis anddifferentiation of HL-60 cells and its possible mechanism.Firstly, the mIκBαcDNA, i.e., IκBαDN cDNA, in which the codingsequences of 1～70 NH2-terminal amino acids containing thephosphorylation sites essential for the activation of NF-κB were deleted, was amplified by reverse transcription-polymerase chain reaction(RT-PCR) from CNE2 cell line of nasopharyngeal carcinoma, whose theIκBαmRNA was abundant. Through digesting with restrictionendonuclease enzyme, recovering by electrophoresis, ligating in vitro andthe using of inter-vector of pCR-Script^TM, the IκBαDN cDNA was thencloned into eukaryotic expression vector of pcDNA3.1(+), or retroviralvector of pCLXSN and pDC316 vector, a shuttle vector for Ad. Theresults indicated that the mutant IκBαcDNA was amplified by RT-PCRfrom CNE2 cells of nasopharyngeal carcinoma and successfully clonedinto the vectors of pcDNA3.1(+), pCLXSN and pDC316 respectively sothat the recombinant vectors of pcDNA-IκBαDN, pCLX-IκBαDN andpDC-IκBαDN were successfully constructed. DNA sequencing resultsfurther showed that the full-length mIκBαcDNA and its normalreading-frame were indeed included, and no any other mutation wasfound in all of the three recombinant vectors of pcDNA-IκBαDN,pCLX-IκBαDN and pDC-IκBαDN.After construction of above recombinant vectors carrying IκBαDNcDNA, the chimeric Ad5F35-IκBαDN AdVec was prepared. Toeffectively transfer IκBαDN cDNA into HL-60 cells, experimentalcondition of Ad5F35 AdVec transfection was firstly optimized. On thebasis of this, the Ad5F35-IκBαDN AdVec was then transduced intoHL-60 cells followed by measuring the Annexin V and CD11b and CD14expression as well as NF-κB-DNA binding activity by fluorescenceactivated cell sorting (FACS) or DNA binding experiment. To assess theAd5F35 AdVec-mediated expression of IκBα, Western blot and real-timeRT-PCR analysis was performed. Meanwhile, the expressions of cIAP-2and xIAP mRNA were also analyzed. The results showed that thepercentage of Annexin V⁺/PI^- and Annexin V⁺/PI⁺ HL-60 cells wasrespectively (22.53±2.999)%, (4.86±1.366)% and (6.08±2.464)% inAd5F35-IκBαDN AdVec-infected or uninfected and Ad5F35-EGFP AdVec-infected HL-60 cells after 48 hours transduction, which had asignificant difference between Ad5F35-IκBαDN AdVec-infected anduninfected HL-60 cells (P＜0.001) or between Ad5F35-IκBαDNAdVec-infected and Ad5F35-EGFP AdVec-infected HL-60 cellsuninfected (0.001＜P＜0.002). But, no significant difference in thepercentage of Annexin V⁺/PI^- and Annexin V⁺/PI⁺ HL-60 cells wasexisted between uninfected and Ad5F35-EGFP AdVec-infected HL-60cells(0.2＜P＜0.5) However, the amount of CD11b- and CD14-expressedpositive cells was a little in Ad5F35-IκBαDN AdVec-infected HL-60cells in 48 hours after transduction, which was equal to that in uninfectedHL-60 cells. Furthermore, NF-κB-DNA binding activity result indicatedthat the average unit of NF-κB-DNA binding activity in Ad5F35-IκBαDN AdVec-infected HL-60 cells, uninfected and Ad5F35-EGFPAdVec-infected HL-60 cells was 0.14, 0.43 and 0.35, respectively.Western blot and real-time RT-PCR results further showed that theexpression of IκBαwas increased with the relative expression level ofIκBαmRNA was 2.51- and 3.16-fold higher then that in uninfected orAd5F35-EGFP AdVec-infected HL-60 cells. In contrast, the relativeabundance of cIAP-2 and xIAP mRNAs was respectively 0.49-, 0.48-,and 0.42-, 0.45-fold lower in HL-60 cells infected with Ad5F35-IκBαDNAdVec compared with that in HL-60 cells uninfected or infected withAd5F35-EGFP AdVec. Taken together these results, we conclude that thechimeric Ad5F35 AdVec can effectively mediate the overexpression oftruncated IκBαin HL-60 cells. Consequentially, NF-κB-DNA bindingactivity was strongly inhibited, and apoptosis was effectively induced.But, the IκBαDN seems not to affect the differentiation of HL-60 cellsinto granulocytes and monocytes, the decreased expression of cIAP-2 andxIAP mRNAs in response to IκBαDN may be involved in IκBαDN-mediated apoptosis.