Cancer Research Annual Meeting 2010 Genetics and Biology of Brain Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Deng, X.
Right arrow Articles by Friedman, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Deng, X.
Right arrow Articles by Friedman, E.
[Cancer Research 66, 4149-4158, April 15, 2006]
© 2006 American Association for Cancer Research

Cell, Tumor, and Stem Cell Biology

The Kinase Mirk/Dyrk1B Mediates Cell Survival in Pancreatic Ductal Adenocarcinoma

Xiaobing Deng, Daina Z. Ewton, Sheena Li, Asghar Naqvi, Stephen E. Mercer, Steve Landas and Eileen Friedman

Department of Pathology, Upstate Medical University, State University of New York, Syracuse, New York

Requests for reprints: Eileen A. Friedman, Pathology Department, Upstate Medical University, 750 East Adams Street, Syracuse, NY 13210. Phone: 315-464-7138; Fax: 315-464-8419; E-mail: friedmae{at}upstate.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Ductal adenocarcinoma of the pancreas is almost uniformly lethal as this cancer is invariably detected at an advanced stage and is resistant to treatment. The serine/threonine kinase Mirk/Dyrk1B has been shown to be antiapoptotic in rhabdomyosarcomas. We have now investigated whether Mirk might mediate survival in another cancer in which Mirk is widely expressed, pancreatic ductal adenocarcinoma. Mirk was an active kinase in each pancreatic cancer cell line where it was detected. Mirk knockdown by RNA interference (RNAi) reduced the clonogenicity of Panc1 pancreatic cancer cells 4-fold and decreased tumor cell number, showing that Mirk mediates survival in these cells. Mirk knockdown by synthetic duplex RNAis in Panc1, AsPc1, and SU86.86 pancreatic cancer cells induced apoptosis and enhanced the apoptosis induced by gemcitibine. Mirk knockdown did not increase the abundance or activation of Akt. However, four of five pancreatic carcinoma cell lines exhibited either elevated Mirk activity or elevated Akt activity, suggesting that pancreatic cancer cells primarily rely on Mirk or Akt for survival signaling. Mirk protein was detected by immunohistochemistry in 25 of 28 cases (89%) of pancreatic ductal adenocarcinoma, with elevated expression in 11 cases (39%). Increased expression of Mirk was seen in pancreatic carcinomas compared with primary cultures of normal ductal epithelium by serial analysis of gene expression and by immunohistochemistry. Thus, Mirk is a survival factor for pancreatic ductal adenocarcinoma. Because knockout of Mirk does not cause embryonic lethality, Mirk is not essential for normal cell growth and may represent a novel therapeutic target. (Cancer Res 2006; 66(8): 4149-58)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Infiltrating ductal adenocarcinoma of the pancreas was the fourth leading cause of cancer-related deaths in the United States in 2004. The annual incidence is similar to the mortality figure because the disease is almost uniformly fatal with a 5-year survival of ~4%. The median survival for pancreatic ductal adenocarcinoma patients is 3 to 6 months because the disease is clinically apparent only at later stages when it is highly resistant to conventional radiation and chemotherapy. Several genes have been associated with the development of pancreatic ductal adenocarcinoma, including mutations in K-ras2, deletion in the INK4A locus encoding the p16 cyclin-dependent kinase (CDK) inhibitor, mutations in p53 and BRAC2, deletions or truncating mutations in smad4/dpc, up-regulation of members of the sonic hedgehog signaling system, increased expression of HER-2/neu, and mutations in mismatch repair genes and in STK11/LKB1 (summarized in refs. 1–3).

Pancreatic ductal adenocarcinomas are so lethal because they are highly resistant to apoptosis induced by chemotherapeutic drugs or by radiation. This inherent antiapoptotic property of pancreatic adenocarcinoma cells could be ascribed in part to the activation of various survival kinases. The serine/threonine kinase Akt, which mediates survival signaling in a number of malignancies, including breast cancer and prostate cancer, was found to be activated in each of eight pancreatic cancer cell lines, and its activation was attributed in part to reduced expression of the lipid phosphatase and tumor suppressor MMAC/PTEN (4), the natural inhibitor of Akt activator, phosphatidylinositol 3 kinase. Activation of phosphatidylinositol 3 kinase itself in pancreatic cancer cells has been attributed to physical interaction with the insulin receptor substrate adaptor molecule (5). However, Akt was found to be activated in only about half of pancreatic cancer specimens (59%: 46 of 78; ref. 6). This study included both localized tumors, which underwent resection, and metastatic tumors, which had been biopsied. In the absence of Akt activation, additional antiapoptotic pathways must function in ductal pancreatic adenocarcinomas. Lung carcinomas and prostate cancers use multiple signaling pathways to block apoptosis, including those using Akt, as well as those activating signal transducers and activators of transcription 3/5 (STAT3/5), extracellular signal-regulated kinase (ERK), c-Jun-NH2-kinase, and p38 mitogen-activated protein kinase (MAPK; refs. 7, 8). Pancreatic ductal adenocarcinomas are also expected to use multiple signaling pathways to mediate tumor cell survival. Unexpectedly, another possible survival kinase for pancreatic ductal adenocarcinoma is glycogen synthase kinase (GSK) 3ß, a kinase inhibited by Akt. Pharmacologic inhibition or genetic depletion of GSK3ß led to decreased pancreatic carcinoma cell proliferation and survival through modulation of nuclear factor-{kappa}B (NF-{kappa}B) activity in some pancreatic ductal adenocarcinoma cell lines (9). Possibly, Akt and GSK3ß mediate antiapoptotic function in different pancreatic ductal adenocarcinomas.

Another survival kinase that may mediate antiapoptotic functions in pancreatic ductal adenocarcinomas is Mirk. Our group and other investigators have shown that the serine/threonine kinase Mirk has antiapoptotic functions in cancer cells in which Mirk is highly expressed and activated, such as rhabdomyosarcoma cells (10), some colon carcinoma cells (11, 12), and HeLa cervical carcinoma cells (13), as well as in terminally differentiating normal skeletal muscle (14, 15). Mirk is expressed at low levels in most normal tissues, including the pancreas. However, Mirk is up-regulated in normal skeletal muscle, so our group defined the function of Mirk in untransformed cells in the C2C12 myoblast system. A large fraction of cycling myoblasts, 20% to 30%, undergo apoptosis when they are placed in differentiation-inducing culture conditions. Mirk was up-regulated during this initial stage of differentiation (14) and inhibited apoptosis unless depleted by RNA interference (RNAi; ref. 15). Mirk is a stress-activated kinase (16) that mediates expression of contractile proteins in differentiating myoblasts (17), but Mirk is not essential for muscle formation or viability in the embryo (18). Mirk is expressed in a class of skeletal muscle stem cells, termed satellite cells (14). Mirk may facilitate survival of satellite cells that give rise to rhabdomyoblasts in regenerating skeletal muscle after injury (10). We noted that Mirk/Dyrk1B was expressed in pancreatic ductal adenocarcinomas, but was not detectable in short-term cultures prepared from normal pancreatic ductal epithelial cells (this article). The capacity of Mirk to inhibit apoptosis provides a compelling rationale for the expression of Mirk in cancer cells, and our initial observation that Mirk was up-regulated in pancreatic ductal adenocarcinoma led us to discover a survival function for Mirk in this cancer.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials. PLUS reagent and LipofectAMINE were from Invitrogen (Carlsbad, CA). Polyvinylidene difluoride transfer paper Immobilon-P was purchased from Millipore (Billerica, MA). All radioactive materials were purchased from Perkin-Elmer (Wellesley, MA), and enhanced chemiluminescence reagents were from Amersham (Piscataway, NJ). Tissue culture medium and sera were obtained from Mediatech (Hampton, NH). Antibody to p27kip1 was from Clontech (Mountain View, CA); antibodies to Akt, phospho-Akt, GSK3ß, and phospho-GSK3ß were from Cell Signaling (Danvers, MA), whereas antibody to ß-tubulin was from Santa Cruz Biotechnology (Santa Cruz, CA). All other reagents were from Sigma (St. Louis, MO).

Cell culture. All pancreatic ductal adenocarcinoma cell lines were obtained from American Type Culture Collection (Manassas, VA) and maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS) and antibiotics. Western blotting was done as described (14).

RNA interference. Both plasmid-based and synthetic RNAi duplexes were used. For plasmid-based studies, each RNAi sequence was inserted into the pSilencer plasmid (Ambion, Austin, TX) and used as described (14, 19). The Si1 RNAi sequence to human Mirk was GACCTACAAGCACATCATT, the Si2 RNAi to Mirk was GCTCTCTGTGGACCTCA, whereas the mutant RNAi sequence was CAGAGCCTACCGATACAG. Panc1 cells (5 x 105) were plated in 60 mm dishes. After 24 hours, cells were transiently transfected with a complex of PLUS reagent, LipofectAMINE (each at 2 µL/µg DNA), 5 µg pSilencer DNA with appropriate insert, and 0.5 µg phosphorylated enhanced green fluorescent protein (pEGFP) for 4 to 6 hours in serum-free medium. Serum was replaced and expression allowed for 24 to 72 hours. Transfection efficiency was determined by visualizing exogenous EGFP using fluorescence microscopy. In colony formation assays, 0.5 µg pCDNA3.1 containing the neomycin resistance gene was cotransfected instead of pEGFP. After 48 hours, cells were trypsinized and replated at single cell density in 100 mm dishes at a series of input densities from 3.5 x 104 to 2.1 x 105 per plate for triplicate measurements per point and selected in G418 sulfate (500 µg/mL) containing medium that was refreshed every 4 days. In some experiments, the blasticidin resistance gene on pCDNA6 was used as a selection marker, with culture medium supplemented with 5 µg/mL blasticidin HCl (Invitrogen).

In synthetic duplex RNAis, duplexes corresponding to mRNA sequences initiated at bases 649 (siA), 828 (siB), and 840 (siC; Invitrogen) were, respectively, directed against sequences within exons 5, 6, and 6. Exon 5 encodes Mirk kinase conserved subdomains II and III, whereas exon 6 encodes Mirk kinase subdomains IV through VII. All of the sequences were unique to Mirk by Basic Local Alignment Search Tool search. Cells were transfected as above, but without any selection markers. A GC-matched negative control duplex RNAi was used. A fluorescent oligonucleotide (BLOCK-iT, Invitrogen) was cotransfected for visual assessment of the efficiency of RNAi transfection, which was >90% in all cases.

Immune complex kinase assay for Mirk activity. Mirk was immunoprecipitated from 500 µg lysate from pancreatic adenocarcinoma cells with either affinity-purified polyclonal antibody to Mirk (M) or nonspecific (ns) preimmune serum (2 µg) as noted, and collected with 20 µL of protein A-agarose beads, and extensively washed. These preparations were incubated for 5 minutes at 30°C with 20 µL kinase buffer [50 mmol/L Tris-HCl (pH 7.5), 10 mmol/L MgCl2, and 0.5 mmol/L DTT], containing 50 µmol/L cold ATP plus 5 µCi [32P-{gamma}]ATP and 1 µg of either purified recombinant glutathione S-transferase–histone deacetylase (GST-HDAC, 51-283 amino acids) protein or purified recombinant p27kip1 as substrate, then analyzed by PAGE and autoradiography.

Patients and tumor specimens. Nineteen archived blocks containing formalin-fixed, paraffin-embedded pancreatic ductal adenocarcinomas from University Hospital from 1996 and 2002 were used. Formalin-fixed, paraffin-embedded sections of nine pancreatic ductal adenocarcinomas from Imgenex (San Diego, CA) were also included in the study. Archived blocks containing formalin-fixed, paraffin-embedded, first-trimester products of conception were also obtained in accordance with institutional review procedures for clinical specimen use.

Immunohistochemistry. Slides (5 µm) were deparafinized. After antigen retrieval with citrate, the endogenous peroxidase activity was blocked by incubation with 0.3% hydrogen peroxide. Slides were incubated for 1 hour with a 1:500 dilution of either preimmune rabbit serum or affinity purified rabbit polyclonal antibody raised to a synthetic peptide corresponding to amino acids 595 to 624 in the unique COOH terminus of human Mirk. The anti-Mirk antibody recognizes the 65/69 kDa Mirk doublet and the 75 kDa splice variant (14) on Western blotting. The slides were incubated with biotinylated secondary antibody and then with an avidin-biotin horseradish peroxidase complex (Vector Labs, Burlingame, CA) and counterstained with hematoxylin. All slides were reviewed by one senior pathologist who verified their designation. The intensity of immunoreactivity was graded by two independent observers, using a scale of 0 to +3, and any focal staining was noted. The visual quantitation was confirmed by quantitation by computer densitometry using the IP LabGel Program. Briefly, representative regions of the tumor sections stained for Mirk protein were photographed, then reduced to grayscale. Areas of stained tissue were outlined and measured for intensity of staining (n = 4). The normalized values for +, ++, and +++ were 6.8 ± 2, 15.9 ± 1.2, and 23.2 ± 3.3, respectively. All of the Mirk staining was cytoplasmic. Nuclei only showed blue counterstaining.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Mirk is an active kinase in pancreatic ductal adenocarcinoma cells. Mirk expression is very low in normal pancreatic tissue, but is up-regulated in several cancers, including colon carcinomas, where Mirk protein was identified by Western blotting (12). Mirk protein was readily detected in the pancreatic cancer cell lines Panc1, SU86.86, BxPc3, AsPc1, Capan1, Capan2, and P10.05 by Western blotting (Fig. 1 ; data not shown). It was possible that Mirk was expressed in pancreatic cancers for some reason unrelated to tumor development. If these were so, we would not expect to find that Mirk was an active kinase in pancreatic ductal adenocarcinoma. Mirk is found within large protein complexes of 670 kDa in vivo (20, 21), which may contain other kinases, such as p38 MAPK, which can immunoprecipitate with Mirk (20). Mirk is activated by phosphorylation by the p38 MAPK kinase MKK3 through a stress-mediated pathway, and activated Mirk within cells can be identified by immune complex kinase assay after immunoprecipitation (16). Thus, it is necessary to carefully control the immunoprecipitation conditions and to use substrates in which the Mirk site of phosphorylation has been mapped to enable the measured kinase activity to be ascribed to Mirk. Mirk phosphorylates p27kip1 only at Ser10 (19) so p27 is a highly selective substrate for Mirk. Active Mirk kinase, capable of phosphorylating p27kip1, was immunoprecipitated from Panc1 and SU86.86 pancreatic adenocarcinoma cells (Fig. 1A). No Mirk was detected in MiaPaCa-2 cells, which served as the negative immunoprecipitation control for kinase activity. Western blotting showed that the three known Mirk splice variants were found in pancreatic adenocarcinoma cell lines Panc 1 and SU86.86, and showed, furthermore, that the affinity-purified antibody used in the Western blots was specific enough to be used for immunoprecipitation and for immunohistochemistry (Fig. 1A, bottom).


Figure 1
View larger version (37K):
[in this window]
[in a new window]
  		Figure 1. Mirk is an active kinase in pancreatic cancer cell lines as shown by immune complex kinase reactions on two Mirk substrates in which the Mirk phosphorylation site had been mapped. A (top), Mirk was immunoprecipitated (IP) from Panc 1 (P1) and SU86.86 (SU) pancreatic ductal adenocarcinoma cell lines with MiaPaCa2 (Mia) cells in which Mirk is not expressed serving as the negative control. Western blotting (WB) was used to show the abundance of immunoprecipitated Mirk compared with the total amount of Mirk seen in cell lysates (top two lanes). The Mirk immunoprecipitates were analyzed for their kinase activity by immunocomplex kinase reactions on recombinant p27kip1 (32P-p27), which was analyzed by SDS-PAGE and autoradiography. Mirk phosphorylates p27kip1 only on Ser10. The total amount of p27 protein in each reaction is shown (bottom). Bottom, affinity-purified antibody to the unique COOH terminus of Mirk recognizes the three known Mirk splice variants in pancreatic adenocarcinoma cell lines. Whole cell lysates from cell lines Panc1 (PC1) and SU86.86 (SU86) cultured in growth medium were analyzed by Western blotting and reblotted for tubulin to show equal loading. Dots at the right, molecular weight markers of 98, 64, 50, and 36 kDa. B, Mirk was immunoprecipitated from AsPc-1 and BxPc-3 cell lysates with either affinity-purified polyclonal antibody to Mirk (M) or nonspecific (ns) preimmune serum. An immune complex kinase assay was done on purified recombinant GST-HDAC5 (1-283 amino acids) protein as substrate, then analyzed by PAGE and autoradiography. Mirk phosphorylates GST-HDAC5 (1-283 amino acids) only on Ser279. To eliminate phosphorylation of HDAC5 by kinases that may associate with Mirk in the immunoprecipitates, phosphorylation was compared on mutant HDAC-S259A/S279A (Mutant Substrate), which is not substrate for Mirk, and the Mirk substrate HDAC5-S259A/S279S. The recombinant substrates had the calcium/calmodulin–dependent protein kinase site of Ser259 mutated to alanine to reduce phosphorylation by this widely expressed kinase. The specificity of these constructs for Mirk is confirmed for these reaction conditions (last two lanes). Purified GST-Mirk phosphorylated the Mirk substrate (Mk), but not the mutant substrate (Mt). C, analysis of the protein abundance of Mirk in the immunoprecipitates in (B) by Western blotting. The phosphorylated IgG heavy chain region of the immunoprecipitates on the autoradiogram is shown. D, Mirk depletion in Panc1 pancreatic carcinoma cells by RNAi. Cells were transfected with pSilencer DNA encoding RNAi to Mirk (either sequences Si1 or Si2), a mutant RNAi (Mt) or vector (Vc) together with an EGFP expression plasmid that was used as a loading and transfer control. One set of transfected cells was analyzed for expression of Mirk and GFP after 3 days by Western blotting (Mirk/GFP 3 days), whereas a second set was cotransfected with an expression plasmid for the neomycin resistance gene instead of EGFP and transfected cells were selected in 400 µg/mL G418 for 3 weeks before analysis of Mirk levels by Western blotting (lowest band, Mirk in colonies). The mean ratio of Mirk abundance to GFP (n = 3) after 3 days in cells transfected with expression plasmids for Si1, Si2, mutant RNAi, or vector was, respectively, 0.4, 0.7, 1.0, and 1.0.

 
Lower levels of Mirk protein were detected in AsPc1 and BxPc3 cells than in Panc1 cells, so Mirk activity was tested in these lines under more stringent conditions. HDAC5 is phosphorylated by Mirk at Ser279 (17), a site that is conserved within all class II HDACs. Mirk phosphorylates only this site in a deletion construct of HDAC5 (amino acids 1-283). To eliminate phosphorylation of HDAC5 by kinases that may associate with Mirk in the immunoprecipitates, phosphorylation was compared on mutant HDAC-S259A/S279A (mutant substrate), which is not a substrate for Mirk, and the Mirk substrate HDAC5-S259A/S279S. The recombinant substrates had the calcium/calmodulin–dependent protein kinase site of Ser259 (22) mutated to alanine to reduce phosphorylation by this widely expressed kinase. The substrate specificity was confirmed for these reaction conditions (Fig. 1B, last two lanes). Purified GST-Mirk phosphorylated the Mirk substrate, but not the mutant substrate.

Initial screening of pancreatic ductal adenocarcinoma cell lines had showed that Mirk was expressed at similar levels in the BxPc3 and AsPc1 cell lines. Mirk was immunoprecipitated from lysates of AsPc1 and BxPc3 cells with either affinity-purified polyclonal antibody to Mirk or nonspecific preimmune serum, and immune complex kinase reactions were done using purified recombinant GST-HDAC5 (1-283 amino acids) protein as substrate, then analyzed by PAGE and autoradiography (Fig. 1B). Examination of the amounts of immunoprecipitated Mirk by Western blotting showed that equivalent amounts of Mirk were found in complexes tested on the Mirk substrate and its mutant counterpart, and that a negligible amount of Mirk was found in the control immunoprecipitates (Fig. 1C). Mirk, after immunoprecipitation from AsPc1 and BxPc3 pancreatic adenocarcinoma cells, phosphorylated the Mirk substrate relative to the mutant substrate 7.2- and 5.5-fold, respectively, after correction for activity immunoprecipitated by the nonimmune serum (Fig. 1B). Therefore, Mirk is an active kinase in these two pancreatic ductal adenocarcinoma cell lines, AsPc1 with mutant K-ras G12D and BxPc3 with wild-type K-ras, as well as in Panc1 and Su86.86 cells. These experiments also indicated that activated mutant K-ras is not essential for Mirk activation in pancreatic ductal adenocarcinoma.

Mirk mediates survival of clonogenic pancreatic ductal adenocarcinoma cells. The biological result of depletion of Mirk by RNAi in pancreatic ductal adenocarcinoma was determined using the pSilencer expression plasmid. Two RNA sequences directed to the Mirk coding sequence within exon 3 were used. Mirk was depleted in Panc1 cells by knockdown with either Si1 or Si2, whereas a mutant RNAi served as the control. After 3 days of treatment, Si1 depleted Mirk protein levels ~60% to 70%, whereas Si2 depleted Mirk on average ~30% compared with mutant RNAi (Fig. 1D, replicate experiments with similar results not shown). These Western blotting studies reflect a large subpopulation of nontransfected cells in which there was no change in Mirk levels. Pancreatic carcinoma cultures contain cells with varying degrees of growth capability. We decided to determine whether depletion of Mirk would affect the most aggressive cells, those capable of growth at single cell density. Panc1 cells were treated with Si1 to deplete Mirk, whereas parallel cultures were treated with the mutant Si. The treated cells were plated at single cell density and cultured for 3 weeks in G418 to select for a cotransfected neomycin resistance plasmid (Fig. 2A ). In this colony formation assay, only ~0.5% of control cells gave rise to colonies. Depletion of Mirk reduced the colony-forming ability of Panc1 cells to 25% of control levels (Fig. 2A). Thus, depletion of Mirk markedly reduced the viability of the most aggressively proliferating cells within the carcinoma cell culture.


Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		Figure 2. Depletion of Mirk in Panc1 cells by RNAi reduced Panc1 pancreatic carcinoma cell viability as measured by colony formation and direct cell counting. A, Mirk knockdown was induced in Panc1 cells by transfection of pSilencer encoding Si1 (Mirk si) or a mutant sequence. Two days posttransfection, cells were plated at single cell density and cultured in medium containing 600 µg/mL G418 for 17 days until individual colonies were discernible. Colonies were fixed and stained in crystal violet (n = 9 for each point). B, Mirk knockdown was induced in Panc1 cells by RNAi by transfection of pSilencer encoding Si1 or Si2 with the control of pSilencer encoding a mutant sequence. Parallel cultures were analyzed for Mirk expression after 3 days and the amount of Mirk protein after Si treatment was compared with the mutant control, with the mutant giving 100% Mirk values, Si2 reducing Mirk values to 70% of control levels, and Si1 reducing Mirk levels to 40% of control levels. Each plasmid was cotransfected with an expression plasmid for the neomycin resistance gene and transfected cells were selected in 400 µg/mL G418 for 3 weeks. Mean colony number is shown (n = 3). The colonies that arose were then lysed on the plate and the amount of protein (µg/mL x 10–2) was determined as a separate measurement of the number of viable cells after Mirk knockdown. Columns, mean; bars, SE (only if >5%). C, Mirk knockdown causes a slow loss in viability. Mirk knockdown was induced in Panc1 cells by RNA inference by transfection of pSilencer encoding Si1 with two controls, pSilencer encoding a mutant sequence, or an empty vector. Each plasmid was cotransfected with an expression plasmid for the blasticidin resistance gene. Transfected cells were selected in 5 µg/mL blasticidin HCl for 0 to 8 days. Cell number was then determined by visual counting using a hematocytometer (n = 5). Columns, mean; bars, SE. One of duplicate experiments is shown with similar results.

 
A similar experiment was done to determine whether there was a linear relationship between the amount of Mirk in Panc1 cells and their capacity for growth. Mirk was depleted by varying amounts by treating Panc1 cells with either Si1 or Si2, with the mutant Si serving as the control (Fig. 1D). A 60% depletion of Mirk by Si1 at the time of cell plating led to a 3-fold decrease in the number of viable colonies after 3 weeks, whereas the smaller 30% knockdown in Mirk levels by Si2 in this experiment was reflected in a 30% decrease in colony number after 3 weeks (Fig. 2B, one of duplicate experiments is shown). The control colonies seemed larger by visual examination than the colonies, which arose from cultures after Mirk knockdown, so the mean total protein concentration in the lysed colonies was also determined (Fig. 2B). The total protein concentration in the colony lysates was 5-fold greater in the mutant RNAi control than in the Si1-treated colonies and 3-fold greater than in the Si-2-treated colonies. These measurements provided an independent confirmation that Mirk knockdown reduced the number of viable cells able to proliferate to form colonies (Fig. 2B). Thus, the greater the depletion of Mirk, the greater the loss in pancreatic cancer cell clonogenicity.

Conversely, growth of pancreatic cancer cell colonies from single cells selected for those cells with elevated Mirk levels. In the experiments shown in Fig. 2A and B, a cotransfected neomycin resistance marker was used to select for Panc1 cells that had been transiently transfected with either Si1 or Si2 and so experienced Mirk knockdown at the time of plating. Depletion of Mirk protein was evident after 2 days and maximal after 3 days. Mirk levels remained depressed for at least 5 days (data not shown). However, colonies did arise after 3 weeks under these selection conditions. We questioned whether colonies that arose from the Si1- and Si2-treated cultures retained low levels of Mirk protein. However, the colonies that grew in G418 selection medium after 3 weeks expressed similar levels of Mirk protein whether they had been treated with Si1, Si2, mutant Si, or the control vector (Fig. 1D lowest band, one of triplicate experiments shown). These studies imply that the cells in which Mirk was depleted after 2 to 5 days of treatment were lost within a few days after plating as single cells. Thus, only the Panc1 cells that expressed Mirk as well as control cells were able to grow to form colonies. The growing cells maintained the neomycin resistance plasmid, but either never received the cotransfected pSilencer plasmids expressing either Si1 or Si2, or quickly lost or functionally inactivated these plasmids. Colony formation from single cells in G418 selection medium selected for the G418 resistance plasmid, but against the Mirk RNAi expression plasmids. Thus, there is a strong selection for elevated Mirk protein levels in proliferating pancreatic cancer cells, consistent with the antiapoptotic functions of Mirk.

Next, the effect of Mirk knockdown on the number of viable and proliferating cells was determined by performing direct cell counting. The more effective Si1 sequence was used to knockdown Mirk using the pSilencer plasmid with a coexpressed blasticidin resistance marker, followed by 8 days of selection in blasticidin-containing medium (Fig. 2C). Six days of treatment with blastocidin killed the nontransfected cells, which comprised approximately two of three of the cells within each treatment group. At this point, the medium was changed and many of the Mirk-depleted cells detached from the plates and were lost. After 2 more days of treatment, the cells treated with the mutant Si or the vector and retaining the blastocidin resistance marker had begun to proliferate, and so these cultures increased in cell number. In contrast, the number of Si1-treated cells continued to decrease, so that after 8 days there were roughly 4-fold as many control cells (Fig. 2C, one of duplicate experiments with similar results shown). Thus, depletion of Mirk causes a slow asynchronous cell death over a period of days, with depletion maximum after 3 days and widespread cell death apparent 5 days after that.

Comparison of the activation of Mirk and Akt in pancreatic ductal adenocarcinoma cells. Deregulation of the serine/threonine kinase Akt has been implicated in the pathogenesis of many human cancers. The strong resistance of pancreatic adenocarcinoma to chemotherapeutic drugs has been ascribed, at least in part, to active Akt (23). Akt was found to be activated in each of eight pancreatic cancer cell lines (4) and in about half of pancreatic cancer specimens (59%: 46 of 78; ref. 6), which included both localized tumors, which underwent resection, and metastatic tumors, which had been biopsied. Another kinase, GSK3ß, was shown to mediate pancreatic cancer cell line proliferation and survival through modulation of NF-{kappa}B activity (9). The abundance of Mirk, Akt activated by phosphorylation at Ser473, total Akt, activated GSK3ß (not shown), and total GSK3ß was measured in a panel of six pancreatic cancer cell lines by Western blotting (Fig. 3A ). Mirk levels are elevated in nontransformed cells in G0-G1 (19) so the abundance of CDK inhibitor p27kip1 was surveyed as a general measure of the fraction of cells in G1. There was no relationship between p27kip1 levels and Mirk levels as Mirk levels were highest in Panc1 and SU86.86 cells and p27 levels were high in SU86.86 cells and low in Panc1 cells (Fig. 3A).


Figure 3
View larger version (52K):
[in this window]
[in a new window]
  		Figure 3. Comparison of Mirk and activated Akt in pancreatic cancer cell lines. A, pancreatic ductal adenocarcinoma cells were cultured for 2 days in growth medium containing 10% FBS and lysates were immunoblotted for Mirk, Akt phosphorylated at Ser473, total Akt, GSK3ß, and p27kip1. Cell lines assayed were AsPc1 (As), CAPAN1 (C1), CAPAN2 (C2), Panc1 (P1), SU86.86 (SU), and MiaPaCa2 (Mia). B, pancreatic ductal adenocarcinoma cells were cultured for 2 days in medium containing 0.5% FBS and lysates were immunoblotted for Mirk, Akt phosphorylated at Ser473, total Akt, and ß-tubulin, as a blotting control. Cell lines assayed were AsPc1, CAPAN1, CAPAN2, Panc1, SU86.86, and MiaPaCa2. C, the abundance and kinase activity of immunoprecipitated Mirk protein on p27kip1 and on HDAC5 were compared in five pancreatic adenocarcinoma cell lines. Data were taken from Fig. 1, and additional data not shown. There was a close relationship between the amount of Mirk protein and the kinase activity of Mirk in these pancreatic cancer cell lines. D, the abundance of Mirk protein was compared with the amount of activated Akt/total Akt in five pancreatic adenocarcinoma cell lines cultured in medium containing 10% FBS, taken from (A). E, the abundance of Mirk protein was compared with the amount of activated Akt/total Akt in five pancreatic adenocarcinoma cell lines cultured in medium containing 0.5% FBS, taken from (B).

 
There seemed to be an inverse relationship between the activation of Akt and the abundance of Mirk in cells cultured in normal growth medium, which contained 10% FBS (Fig. 3A), and, except for the CAPAN1 line, in cells cultured under conditions of serum restriction (Fig. 3B). The abundance of Mirk and Mirk kinase activity were closely correlated in pancreatic cancer cells, indicating that Mirk was consistently activated in pancreatic ductal adenocarcinomas (Fig. 3C). Therefore, Mirk abundance, as a correlate for Mirk activity, was plotted against the amount of activated Akt normalized to the total amount of expressed Akt in each cell line (Fig. 3D). In cells in growth medium, the abundance and activity of Mirk were highest in Panc1 and SU86.86 cells with much lower activities in the AsPc1, CAPAN1, and CAPAN2 cell lines. Akt exhibited the opposite pattern. Akt had the highest activity in AsPc1, CAPAN1, and CAPAN2 cells, with much lower activity in Panc1 and Su86.86 cells (Fig. 3D). Akt can be activated by insulin-like growth factor (IGF) receptor signaling initiated by binding to the IGF in serum, so Mirk and Akt activity were compared in cells cultured in low serum medium. Akt and Mirk displayed a similar pattern of alternate activation, except in CAPAN1 cells under these restricted serum conditions (Fig. 3E). The total amount of GSK3ß and its activation was similar in each line (Fig. 3A; data not shown). Mirk was not expressed in Mia-PaCa2 cells (Fig. 3A), which exhibited elevated GSK3ß activity as a survival kinase (9). These data, taken together, suggest a model in which pancreatic adenocarcinomas that express high levels of Mirk, like Panc1 and SU86.86 cells, may primarily use Mirk as a survival kinase, whereas those cancers that express Mirk at low levels instead use Akt or GSK3ß as survival kinases. This hypothesis was tested in the following studies.

Mirk depletion does not increase abundance or activation of Akt. Endogenous Mirk in Panc1 cells was depleted to 5% to 8% of control levels by treatment for 3 days with each of three synthetic duplex RNAis directed to different regions within the coding sequence of Mirk within exons 5 or 6 (Fig. 4A ). Although Mirk was almost completely depleted in these cultures, the total amount of Akt was not altered, and there was little change in the activation of Akt (Fig. 4A, bottom). Therefore, depletion of Mirk did not activate Akt. These data imply that loss of clonogenicity of Panc1 cells by depletion of Mirk (Fig. 2) occurred under conditions in which Akt was not up-regulated or activated to compensate for loss of Mirk.


Figure 4
View larger version (56K):
[in this window]
[in a new window]
  		Figure 4. Mirk knockdown did not affect Akt, but enhanced drug-induced apoptosis. A, siRNA duplexes corresponding to mRNA sequences initiated at bases 649 (A), 828 (B), and 840 (C) were directed against sequences within exon 5 or 6. Panc1 cells in six-well plates were treated for 3 days with 250 pmol duplex RNAis per well. The control duplex RNAi (Ct) was GC matched to siA. Lysates were analyzed for abundance of Mirk, Akt, and Akt activated by phosphorylation at Ser473 by Western blotting. B, depletion of Mirk in pancreatic ductal adenocarcinoma cells by RNAi enhanced drug-induced apoptosis. Panc1 cells (top), AsPc1 cells (middle), and SU86.86 cells (bottom) in six-well plates were treated for 1 day with 250 pmol duplex RNAis siA and siC per well, then gemcitibine at 100 µmol/L (Panc1) or 200 µmol/L was added for an additional 24 hours before cell lysis. The control duplex RNAi was GC matched to siA. Lysates were analyzed for the abundance of Mirk, cleaved PARP, and tubulin as the blotting control. The ratio of cleaved PARP to tubulin is given below the lanes. C, depletion of Mirk in Panc1 cells by RNAi enhanced drug-induced apoptosis. Mirk knockdown was induced in Panc1 cells by transfection of pSilencer encoding Si1 or a mutant sequence. Cells were cultured in medium supplemented with 0.5% FBS for 2 days to enable Mirk turnover to occur, then treated for another 24 hours with 100 µmol/L gemcitibine (Gem) or 250 ng/mL nocodazole (Noc). The amounts of Mirk, cleaved PARP, cleaved activated caspase-3, and ß-tubulin were determined by Western blotting. Mirk was reduced to 0.27 of control levels. The increase in amount of cleaved caspase-3 and cleaved PARP by depletion of Mirk by RNAi compared with controls averaged 2.1-fold. The amounts of each protein following treatment by Si1 relative to mutant RNAi after 2 to 3 days were quantitated and normalized to tubulin. Columns, mean; bars, SE (only if >5%). D, SAGE using the NIH public database shows that Mirk expression is elevated in pancreatic cancers compared with normal pancreatic ductal epithelium. H126 and HX are two short-term cultures of normal pancreatic epithelial ductal cells that showed no detectable Mirk expression. Panc91 and Panc96 are resected ductal adenocarcinomas of the pancreas, whereas Panc1 and CAPAN1 are established pancreatic adenocarcinoma cell lines. The normal muscle data is the mean of SAGE analysis of biopsies from both old and young subjects.

 
Mirk knockdown by synthetic duplex RNAis enhances response to gemcitibine. We next tested whether depletion of Mirk would induce apoptosis and in addition, render pancreatic ductal adenocarcinoma cells more sensitive to apoptosis induced by gemcitibine, a drug commonly used in treatment of pancreatic ductal adenocarcinoma. A second question was whether pancreatic ductal adenocarcinoma cells with high Mirk levels would be more sensitive to Mirk knockdown than pancreatic ductal adenocarcinoma cells with low Mirk levels. Panc1 cells and SU86.86 cells (high Mirk expressors) and AsPc1 cells (low Mirk expressors; see Fig. 3A) in six-well plates were treated for 1 day with 250 pmol duplex RNAis per well, then gemcitibine was added for an additional 24 hours to half of the cultures before cell lysis (Fig. 4B). The control duplex RNAi was GC matched to siA. Mirk levels, normalized to tubulin levels, were decreased to a mean of 8% of control levels by siA and to a mean of 5% of control levels by siC in Panc1 cells, and to 19% and 15% of control levels, respectively, in Su86.86 cells (Fig. 4B). Knockdown of Mirk by siA and siC reduced the already low Mirk levels in AsPc1 cells to levels undetectable by Western blotting (Fig. 4B, middle). A critical step in apoptosis is the cleavage of poly(ADP-ribose) polymerase (PARP). Depletion of Mirk in the absence of gemcitibine increased apoptosis in each cell line, as assayed by the levels of cleaved PARP normalized to tubulin (Fig. 4B). The amount of apoptosis induced by Mirk knockdown was not related to the amount of Mirk initially expressed. Knock down of Mirk in Panc1 cells with high initial Mirk expression led to a 2- to 3-fold increase in the amount of cleaved PARP, whereas knockdown of Mirk in Su86.86 cells, also with high initial Mirk expression, led to a much greater increase in the apoptotic marker, 11- to 21-fold. In AsPc1 cells, depletion of the low levels of Mirk initially present by only one synthetic duplex RNAi sequence, siC, increased cleaved PARP levels 5-fold. Thus, knockdown of Mirk increased apoptosis in each of the three pancreatic cancer cell lines, and the extent of the increase was unrelated to the amount of Mirk present initially.

Similarly, Mirk knockdown by the more effective RNAi compound siC before addition of gemcitibine increased PARP cleavage relative to drug treatment alone 2- to 3-fold in Panc1, AsPc1, and Su86.86 cells. Mirk knockdown by the less effective siA before drug treatment increased PARP cleavage <2-fold. Thus, depletion of Mirk rendered pancreatic cancer cells with high Mirk expression (Panc1 and Su86.86) or low Mirk expression (AsPc1) more sensitive to gemcitibine, a drug used clinically for pancreatic ductal adenocarcinoma.

Mirk knockdown by a plasmid-based RNAi, directed to a different region of the coding sequence than any of the synthetic duplex RNAis, sensitizes Panc1 cells to apoptosis induced by gemcitibine and nocodazole. The pSilencer expression plasmid encoding the Si1 sequence was used to knock down Mirk, with the pSilencer expression plasmid encoding a mutant sequence as control. Mirk was depleted to 27% to 35% of control levels in Panc1 cells after 3 days of treatment and was maintained at ~50% of control levels for 4 and 5 days in duplicate experiments (data not shown). The 3-fold decrease seen in Mirk protein levels in mass cultures in these experiments is due to low transfection efficiency of Panc1 cells. Therefore, the Western blotting studies reflect a large subpopulation of nontransfected cells in which there was no change in Mirk levels. Cells were cultured in medium supplemented with only 0.5% FBS to induce apoptosis. Critical steps in apoptosis are the activation of caspases as well as the cleavage of PARP. Depletion of Mirk in the subpopulation of transfected cells resulted in a 2-fold increase in the amount of cleaved, activated caspase-3, and cleaved PARP (Fig. 4C). In parallel cultures, cells were also treated with the chemotherapeutic agents nocodazole or gemcitibine after depletion of Mirk. PARP cleavage was increased to 3-fold by Mirk depletion plus gemcitibine or nocodazole, respectively, whereas caspase-3 cleavage was increased 3- to 4-fold by each drug following depletion of Mirk (Fig. 4C). Thus, Mirk knockdown in pancreatic ductal adenocarcinoma enhanced apoptotic cell death and increased the cell killing caused by gemcitibine and nocodazole.

Serial analysis of gene expression. Mirk expression is elevated in pancreatic cancers compared with normal pancreatic ductal epithelium by serial analysis of gene expression (SAGE) using the NIH public database (Fig. 4D). Mirk was expressed at about 75 tags per million in normal skeletal muscle, the normal cell type in which Mirk is most highly expressed (12) and in the pancreatic ductal carcinoma cell line Panc1. Mirk was also expressed in the resected pancreatic ductal adenocarcinomas Panc96 and Panc91, and the pancreatic cancer cell line CAPAN1. The SAGE expression data on Panc1 and CAPAN1 cells (Fig. 4D) paralleled their relative abundance by Western blotting (Fig. 3A). In sharp contrast, Mirk transcripts were not detectable in H126 and HX, two short-term cultures of normal pancreatic ductal epithelial cells. These data were surprising because Mirk was not detected as a differentially expressed gene in microarray experiments by other investigators (24–26). However, Mirk expression is quite low, as expected for a serine/threonine kinase that acts catalytically, so Mirk might be missed in microarray screens that typically identify the more abundant proteins.

Mirk is widely expressed in clinical cases of resected pancreatic ductal adenocarcinoma. Archived blocks containing formalin-fixed, paraffin-embedded pancreatic ductal adenocarcinomas from University Hospital from 1996 and 2002 were obtained. Three sections were cut from each block. One served as the negative control, one was stained with H&E and indicated the tumor tissue to the pathologist, and the third was incubated with affinity-purified antibody to Mirk. A section of skeletal muscle was placed adjacent to each tumor section to serve as an internal positive control for Mirk expression (14). Formalin-fixed, paraffin-embedded sections of additional pancreatic ductal adenocarcinomas from Imgenex were also included in the study. Twenty-five of 28 (89%) of the pancreatic adenocarcinomas expressed Mirk protein (Table 1 ). In 18 of these cases (64%), Mirk protein was present in at least 20% of the cells. In 39% of the tumors (11 of 28), Mirk intensity of staining was ++ to +++ and Mirk was present in at least 20% of the cells.


View this table:
[in this window]
[in a new window]
  		Table 1. Mirk expression in pancreatic ductal adenocarcinoma

 
In earlier studies in nontransformed epithelial cells, fibroblasts, and myoblasts, Mirk was predominately in the nucleus where it helped to mediate G0 arrest by phosphorylating cell cycle regulatory molecules like p27kip1 and cyclin D1 (14, 19, 27). In sharp contrast to this nuclear localization, all of the Mirk stainings in pancreatic ductal adenocarcinoma were cytoplasmic (Fig. 5A and B ). Nuclei in pancreatic ductal adenocarcinoma only showed blue counterstaining (Fig. 5B). Thus, we would expect the growth arrest properties of Mirk to be abrogated in pancreatic cancers by localization of Mirk to the cytoplasm where it would interact with a different set of potential substrates. Mirk also becomes localized in the cytoplasm when it functions as a survival factor in rhabdomyosarcoma cells and in differentiating myotubes (10, 15). Localization to the cytoplasm may have two functions—abrogation of the posttranslational cell cycle arrest function of Mirk and induction of its antiapoptotic functions.


Figure 5
View larger version (120K):
[in this window]
[in a new window]
  		Figure 5. Mirk expression in pancreatic ductal adenocarcinomas and normal pancreatic tissue by immunohistochemistry. A, moderately differentiated pancreatic ductal adenocarcinoma showing strong Mirk expression. Low background staining in fibrotic stroma is seen. B, higher magnification of the center of the tumor section shown in (A). Note that the tumor nuclei exhibit only the blue counterstain and do not express Mirk. C, normal pancreatic duct (center, black arrow) showing no Mirk expression while some nearby normal acini exhibit some focal Mirk expression. D, parallel control slide to (B) made using preimmune serum instead of Mirk antibody. White arrows in (B and D), the same area of the tumor in sequential sections.

 
There was no relationship between the extent of differentiation in the pancreatic ductal adenocarcinoma and Mirk expression. High levels of Mirk protein were seen in poorly differentiated pancreatic ductal adenocarcinoma and in moderately differentiated pancreatic ductal adenocarcinoma expressing Mirk within a duct and adjacent invasive carcinoma (data not shown). A well-differentiated area of pancreatic ductal adenocarcinoma displaying Mirk protein was found adjacent to an area of poorly differentiated adenocarcinoma displaying little Mirk protein. Normal pancreatic ducts were present in five cases, and either no Mirk expression (Fig. 5C), low Mirk expression, or focal expression of Mirk in a few cells was observed in four of these cases. In one case (patient 5), Mirk levels were high in the tumor and high focally (+2 score) in the normal-appearing ducts, perhaps suggestive of early premalignant changes in some of the epithelial cells composing the ducts. In contrast, Mirk protein was observed in some of the normal pancreatic acini in seven of eight cases (Fig. 5C). Acini are not believed to give rise to pancreatic cancer. Thus, by both SAGE analysis and immunohistochemistry, Mirk levels were low to undetectable in most normal pancreatic ductal epithelial cells, the precursors of pancreatic ductal adenocarcinomas. In contrast, Mirk was widely expressed in pancreatic ductal adenocarcinoma, present in 89% of the cases in the current immunohistochemical survey of clinical specimens. The up-regulation of Mirk in pancreatic ductal adenocarcinoma cells and its cytoplasmic localization are consistent with the survival function of Mirk in this cancer.

Mirk localizes predominately in the cytoplasm in fetal pancreatic ductal epithelial cells. Paraffin-embedded sections of first trimester products of conception were analyzed by immunohistochemistry with anti-Mirk COOH-terminal antibody. Mirk had an appreciable expression in the cytoplasm of developing pancreatic ductal epithelial cells with occasional cells (~5%) demonstrating significant nuclear localization (Fig. 6 ). These results were representative of a total of five fetal pancreases examined. Thus, Mirk is a fetal protein whose expression is diminished in normal adult pancreatic ductal epithelium and which is reexpressed in malignant tissue.


Figure 6
View larger version (114K):
[in this window]
[in a new window]
  		Figure 6. Mirk localizes predominately in the cytoplasm in fetal pancreatic ductal epithelial cells. Paraffin-embedded sections of first-trimester products of conception were analyzed by immunohistochemistry with anti-Mirk COOH-terminal antibody (left) compared with H&E staining for parallel sections (right). Original magnifications are indicated.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Inappropriate control of cell survival can lead to the development of cancer. Pancreatic ductal adenocarcinomas are so lethal because they are highly resistant to apoptosis induced by chemotherapeutic drugs or by radiation. Lung carcinomas and prostate cancers use multiple signaling pathways to block apoptosis, including those using the serine/threonine kinase Akt, STAT3/5, and MAP/ERK kinase/ERK (7, 8); thus, it is likely that pancreatic ductal adenocarcinomas also use several survival signaling pathways. We have shown here that the serine/threonine kinase Mirk has a major role in mediating survival of clonogenic pancreatic ductal adenocarcinomas through the antiapoptotic properties of Mirk. Moreover, Mirk is activated in each pancreatic cancer in which it is detected. Mirk was expressed in the majority (89%) of resected pancreatic cancers, with elevated expression in 39%, and in the developing ductal epithelium of each of five fetal pancreases. Unexpectedly, knockdown of Mirk increased apoptosis in AsPc1 cells with low Mirk expression, as well as increasing apoptosis in Panc1 and SU86.86 cells with higher Mirk expression. A second conclusion from these studies is that knockdown of Mirk by synthetic duplex RNAis sensitized each of three pancreatic cancer cell lines to apoptosis induced by gemcitibine. The increase was 2- to 3-fold over drug treatment alone in each of the three cell lines. Thus, depletion of Mirk rendered pancreatic cancer cells with high Mirk expression (Panc1 and Su86.86) or low Mirk expression (AsPc1) more sensitive to gemcitibine, a drug commonly used clinically to treat pancreatic ductal adenocarcinoma.

There was an inverse relationship between the activation of Akt and Mirk in four of five pancreatic cancer cell lines. The strong resistance of pancreatic adenocarcinoma to chemotherapeutic drugs may be caused, in part, by active Akt (23). Akt was found to be activated in each of eight pancreatic cancer cell lines (4) and in about half of pancreatic cancer specimens (59%: 46 of 78; ref. 6). Akt and Mirk are both serine/threonine kinases. Both Akt and Mirk mediate cell survival in part through phosphorylation of p21cip1 within its nuclear localization domain, Akt at T145/S146 and Mirk at S153, forcing the localization of p21 in the cytoplasm (15, 28). Cytoplasmic p21 inactivates proapoptotic signaling by sequestering procaspase-3 and ASK1 and cannot inhibit cell growth (reviewed in ref 29). A p21 construct, phosphomimetic at the Mirk phosphorylation site, was twice as active as wild-type p21 in suppressing conversion of procaspase-3 to active caspase (15). Overexpression of p21cip1 is an early event in the development of pancreatic intraepithelial neoplasias, the precursor lesion to frank carcinoma (30). Thus, Akt and Mirk may both block apoptosis in pancreatic adenocarcinomas through phosphorylation of p21cip1. In addition, both Mirk and Akt2 were identified as survival kinases in HeLa cervical carcinoma cells by a large-scale RNAi approach (13). In this screen of 650 known and putative kinases, apoptosis was measured by DNA fragmentation. Thus, an independent group of investigators has provided data that Mirk mediates tumor cell survival. Possibly, pancreatic cancer cells use both Mirk and Akt to mediate survival in response to chemotherapeutic drugs.

The full spectrum of Mirk substrates is not yet known, but may include other antiapoptotic proteins, such as STAT3, which is constitutively activated in pancreatic ductal adenocarcinomas (31) and the BH3-interacting domain death agonist Bid. Both proteins contain a canonical Mirk phosphorylation site. Another possible role for Mirk in pancreatic ductal adenocarcinoma may be as a transcriptional coactivator of antiapoptotic proteins. Mirk activates the transcription factor MEF2 by blocking the nuclear accumulation of its inhibitor class II histone deacetylases (17). In addition, Mirk phosphorylates and activates the transcription factor hepatocyte nuclear factor 1{alpha} (16), which is expressed in the normal pancreas. Thus, Mirk may mediate pancreatic ductal adenocarcinoma cell survival by transcriptional or posttranslational mechanisms.

Mirk has other functions in myoblast differentiation, which are unlikely to be used in pancreatic cancer cells. When Mirk is localized in the nucleus, Mirk aids in the maintenance of G0-G1 arrest of differentiating myoblasts by posttranslational mechanisms, by destabilizing cyclin D1 by phosphorylation at T288 (27), and by stabilizing the CDK inhibitor p27 by phosphorylation at S10 (19). However, Mirk was localized in the cytoplasm in each clinical case of pancreatic ductal adenocarcinoma examined (this report) so Mirk is unlikely to have these growth arrest properties in pancreatic ductal adenocarcinoma. In fact, depletion of Mirk in Panc1 pancreatic adenocarcinoma cells decreased cell growth (this report), whereas if Mirk mediated arrest in G0, Mirk knockdown would be expected to increase cell cycling. Thus, the major function of Mirk in pancreatic ductal adenocarcinoma is to mediate cell survival, not growth arrest.


   Acknowledgments
 
Grant support: NIH grant RO1 CA67405 (E. Friedman).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 9/ 8/05. Revised 1/17/06. Accepted 2/17/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bardeesy N, DePinho R. Pancreatic cancer biology and genetics. Nat Rev Cancer 2002;2:897–909.[CrossRef][Medline]
  2. Berman D, Karhadkar S, Maitra A, et al. Widespread requirement for Hedgehog ligand stimulation in growth of digestive tract tumours. Nature 2003;425:846–50.[CrossRef][Medline]
  3. Thayer S, di Magliano M, Heiser P, et al. Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 2003;425:851–6.[CrossRef][Medline]
  4. Asano T, Yao Y, Zhu J, Abbruzzese J, Reddy S. The PI3-kinase/Akt signaling pathway is activated due to aberrant Pten expression and targets transcription factors NFB and c-Myc in pancreatic cancer cells. Oncogene 2004;23:8571–80.[CrossRef][Medline]
  5. Asano T, Yao Y, Shin S, McCubrey J, Abbruzzese JL, Reddy SAG. Insulin receptor substrate is a mediator of phosphoinositide 3-kinase activation in quiescent pancreatic cancer cells. Cancer Res 2005;65:9164–8.[Abstract/Free Full Text]
  6. Schlieman M, Fahy B, Ramsamooj R, Breckett L, Bold R. Incidence, mechanism and prognostic value of activated AKT in pancreas cancer. Br J Cancer 2003;89:2110–5.[CrossRef][Medline]
  7. Sordella R, Bell D, Haber D, Settleman J. Gefitinib-sensitizing EGFR mutations in lung cancer activate anti-apoptotic pathways. Science 2004;305:1163–7.[Abstract/Free Full Text]
  8. Uzgare AR, Isaacs JT. Enhanced redundancy in Akt and mitogen-activated protein kinase-induced survival of malignant versus normal prostate epithelial cells. Cancer Res 2004;64:6190–9.[Abstract/Free Full Text]
  9. Ougolkov AV, Fernandez-Zapico ME, Savoy DN, Urrutia RA, Billadeau DD. Glycogen synthase kinase-3{ß} participates in nuclear factor {{kappa}}B-mediated gene transcription and cell survival in pancreatic cancer cells. Cancer Res 2005;65:2076–81.[Abstract/Free Full Text]
  10. Mercer S, Ewton D, Shah S, Naqvi A, Friedman E. Mirk/dyrk1B mediates cell survival in rhabdomyosarcomas. Cancer Res. In press 2006.
  11. Lee K-M, Friedman E. Novel dual-specific tyr-thr/ser kinase cloned from colon carcinoma cells. Proc AACR 1998;39:273.
  12. Lee K, Deng X, Friedman E. Mirk protein kinase mediates serum-free survival of colon cancer cells. Cancer Res 2000;60:3631–7.[Abstract/Free Full Text]
  13. MacKeigan JP, Murphy LO, Blenis J. Sensitized RNAi screen of human kinases and phosphatases identifies new regulators of apoptosis and chemoresistance. Nat Cell Biol 2005;7:591.[CrossRef][Medline]
  14. Deng X, Ewton D, Pawlikowski B, Maimone M, Friedman E. Mirk/dyrk1B is a {rho}-induced kinase active in skeletal muscle differentiation. J Biol Chem 2003;278:41347–54.[Abstract/Free Full Text]
  15. Mercer S, Ewton D, Deng X, Lim S, Mazur T, Friedman E. Mirk/dyrk1B mediates survival during the differentiation of C2C12 myoblasts. J Biol Chem 2005;280:25788–801.[Abstract/Free Full Text]
  16. Lim S, Jin K, Friedman E. Mirk protein kinase is activated by MKK3 and functions as a transcriptional activator of HNF1{alpha}. J Biol Chem 2002;277:25040–6.[Abstract/Free Full Text]
  17. Deng X, Ewton D, Mercer S, Friedman E. Mirk/dyrk1B decreases the nuclear accumulation of class II histone deacetylases during skeletal muscle differentiation. J Biol Chem 2005;280:4894–905.[Abstract/Free Full Text]
  18. Leder S, Czajkowska H, Maenz B, et al. Alternative splicing variants of the protein kinase DYRK1B exhibit distinct patterns of expression and functional properties. Biochem J 2003;372:881–8.[CrossRef][Medline]
  19. Deng X, Mercer S, Shah S, Ewton D, Friedman E. The CDK inhibitor p27kip1 is stabilized in G0 by Mirk/dyrk1B kinase. J Biol Chem 2004;279:22498–504.[Abstract/Free Full Text]
  20. Lim S, Zou Y, Friedman E. The transcriptional activator Mirk/dyrk1B is sequestered by p38{alpha}/ß MAP kinase. J Biol Chem 2002;277:49438–45.[Abstract/Free Full Text]
  21. Zou Y, Lim S, Lee K, Deng X, Friedman E. Mirk/Dyrk1B is an inhibitor of epithelial cell migration which negatively regulated by the Met-adaptor ranBPM. J Biol Chem 2003;278:49573–81.[Abstract/Free Full Text]
  22. McKinsey T, Zhang C, Lu J-r, Olson E. Signal-dependent nuclear export of a histone deacetylase regulates muscle differentiation. Nature 2000;408:106–11.[CrossRef][Medline]
  23. Ng SSW, Tsao M-S, Chow S, Hedley DW. Inhibition of phosphatidylinositide 3-kinase enhances gemcitabine-induced apoptosis in human pancreatic cancer cells. Cancer Res 2000;60:5451–5.[Abstract/Free Full Text]
  24. Iacobuzio-Donahue CA, Ashfaq R, Maitra A, et al. Highly expressed genes in pancreatic ductal adenocarcinomas: a comprehensive characterization and comparison of the transcription profiles obtained from three major technologies. Cancer Res 2003;63:8614–22.[Abstract/Free Full Text]
  25. Logsdon CD, Simeone DM, Binkley C, et al. Molecular profiling of pancreatic adenocarcinoma and chronic pancreatitis identifies multiple genes differentially regulated in pancreatic cancer. Cancer Res 2003;63:2649–57.[Abstract/Free Full Text]
  26. Iacobuzio-Donahue CA, Maitra A, Olsen M, et al. Exploration of global gene expression patterns in pancreatic adenocarcinoma using cDNA microarrays. Am J Pathol 2003;162:1151–62.[Abstract/Free Full Text]
  27. Zou Y, Ewton D, Deng D, Mercer S, Friedman E. Mirk/dyrk1B kinase destabilizes cyclin D1 by phosphorylation at threonine 288. J Biol Chem 2004;279:27790–8.[Abstract/Free Full Text]
  28. Zhou B, Liao Y, Xia W, Lee M-H, Hung M-C. Cytoplasmic localization of p21cip1/waf1 by akt-induced phosphorylation in HER-2/neu-overexpressing cells. Nat Cell Biol 2001;3:245–52.[CrossRef][Medline]
  29. Blagosklonny M. Are p27 and p21 cytoplasmic oncoproteins? Cell Cycle 2002;1:391–3.[Medline]
  30. Biankin AV, Kench JG, Morey AL, et al. Overexpression of p21WAF1/CIP1 is an early event in the development of pancreatic intraepithelial neoplasia. Cancer Res 2001;61:8830–7.[Abstract/Free Full Text]
  31. DeArmond D, Brattain M, Jessup J, et al. Autocrine-mediated ErbB-2 kinase activation of STAT3 is required for growth factor independence of pancreatic cancer cell lines. Oncogene 2003;22:7781–95.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
X. Deng, D. Z. Ewton, and E. Friedman
Mirk/Dyrk1B Maintains the Viability of Quiescent Pancreatic Cancer Cells by Reducing Levels of Reactive Oxygen Species
Cancer Res., April 15, 2009; 69(8): 3317 - 3324.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
O. Bogacheva, O. Bogachev, M. Menon, A. Dev, E. Houde, E. I. Valoret, H. M. Prosser, C. L. Creasy, S. J. Pickering, E. Grau, et al.
DYRK3 Dual-specificity Kinase Attenuates Erythropoiesis during Anemia
J. Biol. Chem., December 26, 2008; 283(52): 36665 - 36675.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
K. Jin, S. Park, D. Z. Ewton, and E. Friedman
The Survival Kinase Mirk/Dyrk1B Is a Downstream Effector of Oncogenic K-ras in Pancreatic Cancer
Cancer Res., August 1, 2007; 67(15): 7247 - 7255.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
R. Kuuselo, K. Savinainen, D. O. Azorsa, G. D. Basu, R. Karhu, S. Tuzmen, S. Mousses, and A. Kallioniemi
Intersex-like (IXL) Is a Cell Survival Regulator in Pancreatic Cancer with 19q13 Amplification
Cancer Res., March 1, 2007; 67(5): 1943 - 1949.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Deng, X.
Right arrow Articles by Friedman, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Deng, X.
Right arrow Articles by Friedman, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online