| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
(PKC) Expression and Constitutive Activation in Gastrointestinal Stromal Tumors (GISTs)
1 Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts; 2 Department of Medicine, Oregon Health & Science University, OHSU Cancer Institute and Portland VA Medical Center, Portland, Oregon; 3 OHSU Cancer Institute and Department of Pathology, Oregon Health & Science University, Portland, Oregon; and 4 Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
(PKC) is a diagnostic marker in GISTs, including those that lack KIT expression and/or contain PDGFRA mutations. PKC is strongly activated in most GISTs and hence may serve, along with KIT/PDGFRA, as a novel therapeutic target. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Protein kinase C (PKC) is a serine/threonine kinase that is transcriptionally up-regulated in GISTs compared with other soft tissue tumors (10
, 11)
. PKC also is the PKC family member that is selectively expressed in the interstitial cell of Cajal lineage (12, 13, 14)
. The PKC family consists of at least 11 related protein kinases that can be divided into three subgroups based on their structural and biochemical properties. PKC, together with PKC
, PKC
, PKC
, and PKCµ, belongs to the so-called novel PKC subgroup, which is characterized by non-calcium-dependent responsiveness to phorbol ester/diacyl-glycerol. In T lymphocytes, PKC is a key signaling molecule in T-cell receptor activation pathways, serving as a positive regulator of cell survival (15, 16, 17, 18)
. PKC inhibition also results in p53-independent cell cycle arrest in various cell types, including mesenchymal (NIH-mouse fibroblast) cells (19)
. These findings suggest that PKC protein expression may be relevant diagnostically and therapeutically in GISTs (20)
. We report here that PKC protein is expressed strongly and is constitutively phosphorylated and enzymatically active in GISTs, irrespective of KIT expression and mutational status. By contrast, PKC protein expression is not detectable, or at most weak, in tumors that are histopathologic mimics of GIST. Our findings demonstrate that PKC is a diagnostic marker and has therapeutic promise for GIST.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
immunoblot analyses were performed in total cell lysates from 53 frozen tumor specimens, including GISTs (n = 20) and potential GIST mimics (n = 33). The GISTs included 15 cases that were KIT positive and 5 that were KIT negative by immunohistochemistry (IHC; DAKO, Carpinteria, CA). Each of the KIT-positive GISTs had genomic KIT mutations, which were located in exon 11 (n = 9), exon 9 (n = 4), exon 13 (n = 1), or exon 17 (n = 1). These KIT-positive and KIT-mutant GISTs were of various primary sites, including gastric (n = 9), small bowel (n = 5), and peritoneal (n = 1). Each of the KIT-negative GISTs had PDGFRA mutations, including three with exon 18 mutations encoding D842V substitutions and two with in-frame exon 12 deletions. Each of these KIT-negative and PDGFRA-mutant GISTs was gastric (n = 5). The diagnosis of GIST in the KIT-negative cases was based on tumor location, typical GIST histopathology, and CD34 immunoreactivity. The histologic GIST mimics, for the immunoblot studies, were leiomyosarcomas (n = 14), leiomyomas (n = 3), malignant peripheral nerve sheath tumors (n = 3), benign schwannomas (n = 2), synovial sarcomas (n = 2), desmoid tumors (n = 2), paragangliomas (n = 3), malignant melanomas (n = 2), and undifferentiated carcinomas (n = 2). In all, 28 of the 33 histologic mimics were from the gastrointestinal tract or from the abdominal or pelvic regions, where primary or metastatic GISTs most often are considered in a differential diagnosis. For the smooth muscle tumors, specifically, the primary sites were esophageal (n = 1), gastric (n = 1), retroperitoneal (n = 10), and pelvic (extrauterine; n = 5).
PKC immunohistochemical analyses were performed in 27 GISTs, which were predominantly KIT-negative cases, and in 40 histologic mimics of GIST, including leiomyosarcomas (n = 20), benign schwannomas (n = 10), and desmoid tumors (n = 10). The KIT-negative GISTs (n = 18) were gastric (n = 11), generalized intra-abdominal (n = 3), omental (n = 2), mesenteric (n = 1), and small bowel (n = 1). These tumors had PDGFRA exon 12 mutations (n = 2), PDGFRA exon 14 mutation (n = 1), PDGFRA exon 18 mutations (n = 10), KIT exon 11 mutation (n = 1), KIT exon 9 mutation (n = 1), or were KIT/PDGFRA-wild-type (n = 3). The KIT-positive GISTs (n = 9) were gastric (n = 3), small bowel (n = 3), omental (n = 1), retroperitoneal (n = 1), and liver metastasis with generalized intra-abdominal involvement (n = 1). The leiomyosarcomas were retroperitoneal (n = 14), uterine (n = 2), and other (n = 4). The desmoid tumors were small bowel, mesenteric, and retroperitoneal (1 of each) and other (n = 7). The schwannomas were retroperitoneal (n = 2) and other (n = 8).
Cell Lines and Reagents.
GIST882 is a primary human GIST cell line with an activating homozygous missense mutation in KIT exon 13, encoding a K642E mutant KIT oncoprotein (21)
. The Jurkat T-cell acute lymphoblastic leukemia cell line was obtained from American Type Culture Collection (Manassas, VA). SARC152 is a primary human cell line established from a high-grade spindle cell sarcoma. Normal thymus protein lysate was from BD Biosciences Clontech (Palo Alto, CA), and normal rat cerebrum protein lysate was from BD Transduction Labs (Lexington, KY).
Antibodies.
Antibodies for total PKC were goat polyclonal sc-1875 (Santa Cruz Biotechnology, Santa Cruz, CA) and mouse monoclonal clone 27 (catalogue no. 610089; BD Biosciences PharMingen, San Diego, CA). Antibodies for phosphorylated PKC were polyclonal rabbit to phospho-Thr538 and phospho-Ser676 (Cell Signaling, Beverly, MA). Incubating Western blots of GIST882 whole cell lysates with calf intestinal alkaline phosphatase was done to validate specificity for phospho-PKC. Other antibodies included monoclonal mouse to PKC, PKC (BD Biosciences PharMingen), and actin (Sigma-Aldrich, St. Louis, MO); polyclonal rabbit to KIT (DAKO), phospho-PDGFRA Y754 (Santa Cruz Biotechnology), and phosphatidylinositol 3'-kinase (Upstate Biotechnology, Lake Placid, NY); and polyclonal goat to myelin basic protein (Santa Cruz Biotechnology).
Western Blot Analysis.
Whole cell lysates of frozen tumors were prepared by mincing the specimens in ice-cold lysis buffer [1% NP40, 50 mM Tris HCl (pH 8.0), 100 mM sodium fluoride, 30 mM sodium PPI, 2 mM sodium molybdate, 5 mM EDTA, 2 mM sodium orthovanadate containing 10 µg/ml aprotinin, 10 µg/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, and additional sodium orthovanadate 2 mM, followed by homogenization with a Tissue Tearor (BioSpec, Bartlesville, OK) and clearing by centrifugation. Protein concentrations were determined with the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). Electrophoresis and immunoblot analysis were performed as described previously (4)
. Blot immunostains were detected by enhanced chemiluminescence (Amersham, Piscataway, NJ), and chemiluminescence signals were captured and quantified using a FUJI LAS1000plus system with Science Lab 2001 ImageGauge 4.0 software (Fujifilm Medical Systems, Stamford, CT).
Immunohistochemistry.
PKC IHC was performed using the mouse monoclonal antibody, clone 27 (catologue no. 610089; BD Biosciences PharMingen), at a 1:100 dilution with 30 min microwave antigen retrieval in 10 mM sodium citrate buffer (pH = 6.0). The EnVision Plus system (DAKO) was used for detection.
In Vitro Kinase Assay.
Endogenous PKC was immunoprecipitated from lysates of serum-starved GIST882 (1 mg of protein) using a goat polyclonal anti-PKC (Santa Cruz Biotechnology) and protein A-Sepharose beads. After overnight incubation and washing in lysis buffer, immunoprecipitates were resuspended in kinase buffer containing 
-32P-ATP (5 µCi per reaction) and myelin basic protein (1 µg per reaction) as substrate. Reactions were incubated for 30 min at 30°C with gentle shaking, then subjected to SDS-PAGE, transferred to nitrocellulose, and developed by autoradiography. The membrane subsequently was stained with polyclonal antibodies to PKC and myelin basic protein.
|
RESULTS AND DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Is Strongly Expressed and Phosphorylated in GISTs, Regardless of KIT or PDGFRA Mutational Status. protein expression in each of 20 GISTs, irrespective of KIT (n = 15) or PDGFRA (n = 5) mutation type (Fig. 1)
. PKC expression was 
50% lower in the PDGFRA-mutant GISTs than in most KIT-mutant GISTs (Fig. 1)
. The relative levels of PKC expression were identical when detected with a goat polyclonal antibody (Fig. 1)
or with a monoclonal mouse antibody (data not shown). Phosphorylation of PKC T538 and S676, which is requisite for PKC activation, was demonstrated in each of the 20 primary GISTs (Fig. 1)
. PKC phosphorylation levels in KIT and PDGFRA mutant GISTs generally paralleled the expression of total PKC. Therefore, per molecule, phosphorylation of PKC was similar in all of the GISTs and was independent of the KIT and PDGFRA mutational mechanisms.
|
expression and phosphorylation were higher in the GISTs than in human thymus and Jurkat T-cell acute lymphoblastic leukemia cells (Fig. 2A)
. This finding is notable because high levels of PKC expression are restricted to few cell lineages, among which particularly high levels have been reported in thymus and Jurkat cells (22
, 23)
. Similarly, PKC phosphorylation was substantially higher in GISTs than in normal thymus or Jurkat cells (Fig. 2A)
. PKC and PKC expression were evaluated in GISTs and cerebrum (positive control), given that these PKC isoforms could cross-react with the phospho-PKC antibodies. However, neither PKC nor PKC was expressed in GISTs (Fig. 2B)
, indicating that the phospho-PKC immunoblot stains were highly specific.
|
activity was evaluated further by in vitro kinase assay in lysates from serum-starved GIST882 cells (Fig. 3)
. Substantial PKC activity as manifested by myelin basic protein phosphorylation was demonstrated in PKC immunoprecipitates from GIST882 but not in the non-GIST spindle cell sarcoma cell line (SAR152). These findings corroborate constitutive PKC activation in GIST.
|
Expression in Histopathologic Mimics of GIST Is Weak to Nondetectable. at levels comparable with those in GISTs (Fig. 4A)
. There was no demonstrable PKC expression in any of 14 leiomyosarcomas, 2 benign schwannomas, 3 paragangliomas, 2 monophasic synovial sarcomas, 2 desmoid tumors, or 2 undifferentiated carcinomas (Fig. 4A)
. There was weak PKC expression (at levels <5% of those in GISTs) in one of three leiomyomas, two of three malignant peripheral nerve sheath tumors, and each of two malignant melanomas (Fig. 4, B and C)
. It is unclear whether the PKC expression in the five non-GIST primary tumors reflects weak, or focal, expression in the neoplastic cells, or perhaps expression in admixed reactive T lymphocytes. Identical results were obtained by immunoblot analysis with a mouse monoclonal antibody to PKC (data not shown). These studies demonstrate that PKC expression generally is at least 20-fold higher in GISTs than in histologic mimics. Importantly, PKC expression was not detected in any leiomyosarcomas, which are the most common differential diagnosis with malignant GIST.
|
expression was evaluated further by IHC in conventional paraffin sections. IHC with the monoclonal PKC antibody demonstrated cytoplasmic staining in 20 of 27 GISTs (72%; Fig. 5A
). PKC expression was shown in 13 of 18 (72%) KIT-negative GISTs, including 11 of 13 cases with PDGFRA mutations, and in 7 of 9 (77%) KIT-positive GISTs. IHC staining intensity varied markedly among the GISTs, in keeping with the differing levels of PKC expression seen by immunoblot analysis, but PKC staining was diffuse in most of the positive cases. By contrast, there was IHC demonstration of PKC expression in only 2 of 20 leiomyosarcomas (one with moderate and diffuse staining and the other with weak and focal staining involving <1% of cells; Fig. 5B
). One of 10 benign schwannomas also was positive for PKC IHC (weak cytoplasmic staining in <5% of cells). All of the 10 desmoid tumors were negative. Correlates with immunoblot analysis (not shown) revealed that IHC with the monoclonal PKC antibody was highly specific but less sensitive than immunoblot analysis. By contrast, IHC with the goat polyclonal PKC antibody (Santa Cruz Biotechnology; data not shown) was highly sensitive but not specific, being positive in several of the non-GIST tumors with no PKC expression by immunoblot analysis. These studies show that a strongly positive PKC IHC result with the BD Biosciences PharMingen monoclonal antibody may be useful to confirm a GIST diagnosis, whereas a negative result is uninformative. Because PKC expression can be demonstrated readily in most GISTs by immunoblot analysis, it is likely that more suitable PKC antibodies, with greater sensitivity and specificity in paraffin section IHC, can be developed.
|
is expressed strongly in GISTs but not in leiomyosarcoma or in other tumors that are histopathologically similar to GIST. Therefore, PKC joins the newly described protein DOG1 as a diagnostic marker of particular relevance in KIT-negative GISTs (24)
. PKC is expressed and phosphorylated in all of the GISTs, irrespective of their KIT or PDGFRA mutational status, demonstrating that PKC is a diagnostic marker and potential therapeutic target for GISTs. It is yet unclear whether PKC activity in GISTs depends on the constitutively activated KIT and PDGFRA signaling pathways in these tumors or whether PKC is activated by mechanisms independent of the KIT and PDGFRA oncogenic signals. However, there is reason to hypothesize that KIT and PDGFRA signaling may be relevant for PKC activation. In particular, KIT- and PDGFRA-mediated activation of phosphatidylinositol 3'-kinase and phospholipase C may generate the PKC cofactors phosphatidylserine and diacylglycerol, respectively (18)
, and thereby modulate PKC function. Although selective inhibition of KIT and/or PDGFRA with agents such as imatinib mesylate has dramatically improved the outcomes of patients with GIST (6)
, resistance to single agent imatinib has been noted to develop over time (25)
. Other kinase inhibitors, such as the multikinase inhibitor SU11248, have shown activity in patients with GIST resistant to imatinib (6)
. However, new therapeutic targets are needed to optimize the care of patients with this life-threatening disease. The data presented here suggest that PKC has many of the attributes that would characterize a GIST-specific target; therefore, it would be worthwhile to test drugs targeting this kinase in patients with imatinib-resistant GIST.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
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.
Requests for reprints: Anette Duensing, University of Pittsburgh Cancer Institute, Hillman Cancer Center, Research Pavilion, Suite 1.8, 5117 Centre Avenue, Pittsburgh, PA 15213. Phone: 412-623-5870; Fax: 412-623-7715; E-mail: aduensin{at}pitt.edu; or Jonathan A. Fletcher, Department of Pathology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. Phone: 617-732-5152; Fax: 617-278-6913; E-mail: jfletcher{at}partners.org
Received 2/17/04. Revised 4/14/04. Accepted 5/20/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
gene in gastrointestinal stromal tumors. Gastroenterology, 125: 660-7, 2003.[CrossRef][Medline]
immunoreactivity to interstitial cells of Cajal in guinea-pig gastrointestinal tract. Neurogastroenterol Motil, 15: 139-47, 2003.[CrossRef][Medline]
during T-cell activation. Nature, 385: 83-6, 1997.[CrossRef][Medline]
and epsilon promote T-cell survival by a rsk-dependent phosphorylation and inactivation of BAD. J Biol Chem, 275: 37246-50, 2000. mediates a selective T cell survival signal via phosphorylation of BAD. J Immunol, 166: 5955-63, 2001. (PKCè): it’s all about location, location, location. Immunol Rev, 192: 53-63, 2003.[CrossRef][Medline]
and isoforms. J Biol Chem, 278: 39782-93, 2003. (PKCè), a potential drug target for therapeutic intervention with human T cell leukemias. Curr Cancer Drug Targets, 2: 125-37, 2002.[CrossRef][Medline]
, a novel member of the protein kinase C (PKC) gene family expressed predominantly in hematopoietic cells. J Biol Chem, 268: 4997-5004, 1993.. Eur J Biochem, 225: 195-203, 1994.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Trojan, M. Montemurro, M. Kamel, and G. Kristiansen Successful PDGFR-{alpha}/{beta} targeting with imatinib in uterine sarcoma Ann. Onc., November 1, 2009; 20(11): 1898 - 1899. [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Demetri, M. C. Heinrich, J. A. Fletcher, C. D.M. Fletcher, A. D. Van den Abbeele, C. L. Corless, C. R. Antonescu, S. George, J. A. Morgan, M. H. Chen, et al. Molecular Target Modulation, Imaging, and Clinical Evaluation of Gastrointestinal Stromal Tumor Patients Treated with Sunitinib Malate after Imatinib Failure Clin. Cancer Res., September 15, 2009; 15(18): 5902 - 5909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Heinrich, S. George, S. Bauer, and J. A. Fletcher Optimizing the Treatment of Gastrointestinal Stromal Tumors: The Roles of Tumor Genotyping, Imatinib Blood-level Testing, and New Therapeutic Strategies ASCO Educational Book, January 1, 2009; 2009(1): 694 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H E Lee, M A Kim, H S Lee, B L Lee, and W H Kim Characteristics of KIT-negative gastrointestinal stromal tumours and diagnostic utility of protein kinase C theta immunostaining J. Clin. Pathol., June 1, 2008; 61(6): 722 - 729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Heinrich, C. L. Corless, C. D. Blanke, G. D. Demetri, H. Joensuu, P. J. Roberts, B. L. Eisenberg, M. von Mehren, C. D.M. Fletcher, K. Sandau, et al. Molecular Correlates of Imatinib Resistance in Gastrointestinal Stromal Tumors J. Clin. Oncol., October 10, 2006; 24(29): 4764 - 4774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L Tornillo and L M Terracciano An update on molecular genetics of gastrointestinal stromal tumours. J. Clin. Pathol., June 1, 2006; 59(6): 557 - 563. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
