We have shown recently (B. A. Yoshida et al., Cancer Res., 59: 5483–5487) that mitogen-activated protein kinase kinase 4 (MKK4) can suppress AT6.1 rat prostate cancer metastases in vivo. Evaluation of the expression of components of the MKK4 signaling cascade showed a loss or down-regulation of expression of MKK4 or c-Jun, a downstream mediator of MKK4, in six of eight human prostate cancer cell lines. Given these findings, we next assessed whether MKK4 dysregulation occurs during the development of clinical prostate cancer. Immunohistochemical studies showed high levels of MKK4 expression in the epithelial but not the stromal compartment of normal prostatic tissues. In neoplastic tissues, a statistically significant, direct, inverse relationship between Gleason pattern and MKK4 was established. These results demonstrate that MKK4 protein is consistently down-regulated during prostate cancer progression and support a role for dysregulation of its signaling cascade in clinical disease. To test the possibility that down-regulation of MKK4 protein is the result of allelic loss, metastatic prostate cancer lesions were examined for loss of heterozygosity (LOH) within the MKK4 locus (D17S969). These studies showed a 31% (5 of 16) LOH of MKK4 that is not associated with coding region mutations, which suggests that the nucleotide sequence of the gene in the remaining allele is infrequently mutated.

Despite advances in prevention and early detection, nearly 20% of men with prostate cancer initially present with clinical evidence of metastases (1, 2, 3, 4). Although refinements in surgical technique and improvements in radiotherapy have improved the cure rates of men with localized prostate cancer, recurrence rates after radical prostatectomy or radiotherapy approach 25–50% (5, 6, 7, 8). Hence, the majority of these men eventually develop metastatic disease (5). It is estimated that 31,900 American men will die this year from metastatic prostate cancer and many more will suffer from the associated morbidity (9, 10).

Metastasis-suppressor genes have been shown to suppress the growth of metastases without affecting the growth of the primary tumor (11). Thus, these genes are distinct from tumor-suppressor genes, which suppress primary tumor growth. There is growing evidence that the loss of metastasis-suppressor gene function is an important event during the development of clinically significant metastases. Recently, we have identified the MAPK4 kinase 4/c-Jun NH2-terminal kinase-activating kinase/stress-activated protein/Erk kinase 1 (MKK4/JNKK1/SEK1, hereafter referred to as MKK4) as a metastasis-suppressor gene encoded by human chromosome 17p11.2 (12). Transfection of highly metastatic AT6.1 Dunning rat prostate cancer cells with MKK4 cDNA reduces the number of overt metastases by ∼80% compared with parental AT6.1 cells and transfection controls, without affecting the growth rate of the primary tumors (12). On the basis of these studies, we hypothesized that inactivation of MKK4 or components of its signaling pathway is involved in human prostate cancer progression and metastasis.

To begin to test this hypothesis, we evaluated the expression of MKK4 and components of its signaling cascade in eight human prostate cancer cell lines, all originally derived from metastatic sites. Next, prostate cancer samples obtained from patients undergoing radical prostatectomy were assessed by immunohistochemistry for MKK4 expression. In the simplest scenario, if MKK4 inactivation plays a role in the development of metastasis, then the level of MKK4 expression in human prostate cancers should be decreased with increasing Gleason grade, an established predictor of metastatic propensity. Finally, to determine whether allelic loss might account for decreased expression of MKK4, 24 human metastatic prostate cancer samples were assessed for LOH using four polymorphic markers in the region of MKK4. The coding regions from selected samples with LOH were sequenced to determine whether MKK4 mutations might account for decreased protein expression.

Cell Lines.

Du145 and TsuPr1 (gifts of Dr. J. T. Isaacs, The Johns Hopkins University, Baltimore, MD), LNCaP and C4–2 (gifts of Dr. L. Chung, University of Virginia, Charlottesville, VA), and PC-3 (gift of Dr. M. Cher, Wayne State University, Detroit, MI) cells were grown in standard DMEM supplemented with 10% FCS. The pancreatic cell line, ASPC-1, which contains a homozygous deletion from D17S969 to MKK4 exon C(13), was obtained from American Type Culture Collection and grown in DMEM supplemented with 20% FCS and 1 mm HEPES. F9 mouse teratocarcinoma cells (gift of Dr. A. Lin, The University of Chicago, Chicago, IL), which have low to undetectable endogenous levels of c-Jun, were grown in 50/50 F12-DMEM supplemented with 10% FCS. The MDA PCa2A and MDA PCa2B cell lines (gift of Dr. N. Navone, M. D. Anderson Cancer Center, Houston, TX) were grown on fibronectin-coated plates BRFF with BRFF-Human Prostate Cancer 1 (HPC1) growth medium supplemented with 10% FCS. Primary prostate epithelial cells were established by treating the fresh prostate tissue in DMEM supplemented with 2 mg/ml collagenase type 1, 10% FCS, 1% Cipro, and 1% penicillin/streptomycin overnight in a 37°C shaker. The cells were then plated on Falcon Primeria tissue culture plates in BRFF medium. All of the cells were grown at 37°C in an atmosphere of 5% CO2.

Immunoblotting.

Cell lines were grown to 80% confluence. Cell lysates were prepared in boiling lysis buffer (100 mm Tris 7.5, 1% SDS, and 100 mm NaVO4), and the protein was quantitated using bicinchoinic assay reduction (Pierce). Thirty μg of total protein from each cell line was subjected to SDS-PAGE (12.5% acrylamide) and transferred onto nitrocellulose membrane. The polyclonal antibodies (Santa Cruz Biotechnology, Inc.) and working dilutions used for immunoblotting were as follows: MKK4 (sc-964, 1:5000), MKK3 (sc-959, 1:10,000), MKK7 (sc-7104, 1:500), MEKK1 (sc-252, 1:1000), JNK1 (sc-571, 1:50,000), p38 (sc-7149, 1:10,000), and c-Jun (sc-1694, 1:5000); cytokeratins 5 and 8 were detected using an antimouse monoclonal antibody (sc-8021, 1:1000).

Membranes were then incubated with a horseradish peroxidase-conjugated IgG secondary antibody and detected using enhanced (Femto) chemilluminescence (Pierce). All of membranes were subsequently reprobed for actin as a loading control (Oncogene Research Products; CP01). Total protein lysates from ASPC-1 cells (30 μg), which have a homozygous deletion of MKK4, and rat brain (3 μg), which expresses high endogenous levels of MKK4, served as negative and positive controls, respectively. Mouse F9 teratocarcinoma cells, which have low endogenous levels of c-Jun, served as a negative control for c-Jun (14). Blocking peptides to MKK3, MKK4, MKK7, and MEKK1 (sc-959P, sc-964P, sc-7104P, and sc-252P, respectively) antibodies were used according to the manufacturer’s instructions to confirm antibody specificity.

Selection of Human Specimens for Immunohistochemistry.

In accordance with a protocol approved by the University of Chicago Institutional Review Board, 69 prostate cancer patient specimens with Gleason patterns 3–5, obtained during radical prostatectomy, were identified from the Department of Pathology database. All of the patient information and linkers were unattainable. Thirty-seven of these patient specimens had multiple tumor loci totaling 112 lesions. The appropriate paraffin blocks, containing tumor and normal prostate tissue, were identified and cut at 5-μm thickness onto statically charged microscope slides for immunohistochemical staining.

Immunohistochemistry.

The slides were treated with xylene and rehydrated by sequential incubations in absolute ethanol, 95% ethanol, and water. After rehydration, the slides were placed in a citrate buffer solution (1.8 mm citric acid, 8.2 mm sodium citrate) at 95°C for 10 min. The slides were then fixed in acetone (5 min) and 0.2% picric acid/2% paraformaldehyde (10 min), and washed in a series of solutions: PBS; 85% ethanol/1.5% PVP; PBS/1.5% PVP; 85% ethanol/0.05% sodium borohydrate/1.5% PVP/PBS; PBS/1.5% PVP; and PBS/1.5% PVP/0.1% gelatin. Endogenous peroxidase was quenched with 3% hydrogen peroxide in methanol for 12 min. Endogenous avidin and biotin were blocked using the Vector Laboratories avidin/biotin blocking kit.

The slides are incubated overnight at 4°C with a primary polyclonal antibody (rabbit anti-MKK4/MEK4; Santa Cruz Biotechnology, Inc.) diluted to 5 μg/ml per manufacturer’s instructions in protein-blocking solution (PBS containing 5% horse serum and 1% goat serum). The slides were incubated for 3 h at 37°C with the secondary antibody (goat antirabbit biotin conjugated IgG; Santa Cruz Biotechnology, Inc.) diluted 3 μg/ml in protein-blocking solution. The regions of antibody binding were visualized using a 3,3′-diaminobenzidine (DAB) peroxidase substrate kit (Vector Laboratories). The slides were counterstained with Fast Green dye (Fisher Scientific).

Several different immunohistochemical controls were performed to verify the sensitivity and specificity of our detection system. ASPC-1 cells, which have a homozygous deletion of the MKK4 gene, were grown on Lab-tek chamber slides (Nalgene) and were used as a negative control, whereas MDA2b prostate cancer cells, which express MKK4 as detected by immunoblotting, were prepared in parallel and were used as a positive control. In addition, sections of normal tissues (e.g., human and rat testes, and human prostate) were used as positive controls. The specificity of the primary antibody used for immunohistochemistry was confirmed by: (a) immunoblotting; (b) the inclusion of isotype controls (using whole rabbit IgG in lieu of the primary antibody); and (c) epitope controls (using the primary antibody preincubated overnight with a 5-fold excess of the peptide; Santa Cruz Biotechnology, Inc.). These controls are summarized in (Table 1).

Evaluation of Clinical Samples.

A semiquantitative evaluation scale was developed for the assessment of MKK4 expression in the clinical samples based on the intensity of staining of normal prostate tissue. In normal prostatic tissue, 100% of the epithelial cells in 16 independent samples showed strong, MKK4-specific expression. In evaluating cancer samples, we used adjacent normal epithelial regions within the slide, which stained for MKK4 expression, as internal controls for evaluating the staining in the tumor region.

To quantitate the staining of the tumor, a numerical scale of “0,” “1,” “2,” and “3” was used, with “3” representing the staining intensity equivalent to the normal histological regions on the same slide and “0” representing no staining. Two independent pathologists (X.Y.) evaluated all of the samples. The average score was taken in the cases in which the scores assigned by the two examiners differed.

LOH Analysis and Mutation Screening.

The tissue samples used for the LOH analysis were collected during rapid autopsies performed on patients with metastatic prostate cancer,. courtesy of Dr. M. Rubin (University of Michigan, Ann Arbor, MI; Ref. 15) and Dr. Tomo Ichikawa (Chiba University, Chiba, Japan). Metastatic lesions and the corresponding prostate tissue were sampled. Normal human and rat AT6.1 DNA were used as positive and negative controls, respectively. Primers to βactin were used to control for the quantity of loading and the integrity of the genomic DNA. PCR primers in the region of MKK4 (D17S954, D17S1303, D17S969, and D17S947) and p53 (D17S796 and D17S938) were purchased from Research Genetics.

Forward primer was end-labeled with (γ-32P)dATP (3000 Ci/mmol; Amersham) using 10 units of T4 kinase. Template genomic DNA (50–100 ng) was amplified by PCR in 10 mm Tris-HCl, 1.5 mm MgCl2, 0.025 units/μl Taq DNA polymerase 100 nm each dNTP, 0.5 μm labeled forward primer, and 0.5 μm unlabeled reverse primer. The PCR conditions included an initial denaturation step at 95°C (5 min) followed by 25 cycles of amplification (1 min at 95°C, 1 min at 57°C, and 1 min at 72°C).

Amplified PCR products were diluted 1:1 with loading dye containing 90% formamide, 10 mm EDTA, 0.05% bromphenol blue, and 0.05% xylene cyanol. The diluted products were denatured at 95°C for 5 min, chilled on ice, separated on 7% polyacrylamide urea gel using electrophoresis at 40 watts, and visualized by autoradiography. Two samples with LOH at the D17S969 locus were further analyzed for specific mutations in MKK4. DNA sequencing and mutation screening were performed for exons B–K, as described by Teng et al.(13).

Statistical Analysis.

Analysis of the association between MKK4 staining score and Gleason pattern was performed using the generalized estimating equations approach of Zeger and Liang (16). This method was necessary because many specimens included multiple regions that varied in Gleason pattern, along with their corresponding staining levels. The generalized estimating equations approach allows for potential correlation among the multiple measurements obtained from a single specimen. Mean staining levels and their SEs were derived based on an identity link function and assuming an exchangeable correlation structure for the “working” correlation matrix. The significance of differences among Gleason patterns was determined using Wald-type tests, in which the difference divided by its SE is referred to the tables of the standard normal distribution.

Analysis of the JNK/SAPK Signaling Cascade in Human Prostate Cancer Cell Lines.

To date, three major MAPK modules have been characterized: ERK, JNK (also known as SAPK), and p38. The JNK and p38 MAPK cascades are subject to regulation by MKK4. The JNK signaling cascade includes: MKK4, MKK7, and MEKK1 (Fig. 1,A; Refs. 17, 18). In response to extracellular stimuli, MEKK1 is a strong activator of MKK4 and MKK7, and a weaker activator of MKK3 and MKK6 (Fig. 1,A; Ref. 19). The p38 signaling cascade includes: MKK3, MKK4, and MKK6 (Fig. 1 A).

We have recently shown that ectopic expression MKK4 suppresses the number of AT6.1 metastases by >80% (12). On the basis of this observation, we hypothesized that inactivation of MKK4 or its signaling cascade is required for the establishment of human prostate cancer metastases. If this is correct, we predicted that expression of MKK4 or component(s) of its signaling pathway would correlate with the acquisition of metastatic ability. To test this, we first assessed the expression of MKK4 in several human prostate cancer cell lines (Fig. 1,B). Rat brain and short term normal human prostate epithelial cultures were used as positive controls, and the ASPC-1 cells, which do not express MKK4 because of a homozygous deletion from D17S969 to exon C, served as a negative control. DU-145, Du-Pro, and PC-3 cells expressed undetectable levels of MKK4. In contrast, MKK4 expression in LNCaP, MDA PCa 2a and 2b, TsuPr1, and C4–2 cells was comparable with that of primary prostate epithelial cells (Fig. 1 B). Additional studies were conducted to test the possibility that cell lines might have alterations in additional components of the MKK4 signaling cascade.

Examination of six well-established components of the MKK4 signaling module showed that all of the cell lines expressed levels of MEKK1, MKK3, MKK7, JNK1, and p38 similar to those in normal prostate epithelial cells. Representative data for JNK1 and p38 are shown in Fig. 1,C. Interestingly, the MDA PCa 2a and 2b cell lines, which were derived from bone metastases, were negative for c-Jun expression (Fig. 1,C). In addition, C4–2, a derivative of the LNCaP cell line that is tumorigenic and metastatic in immunodeficient animals, showed a very low level of expression of c-Jun (Fig. 1 C). F9 cells, which express low endogenous levels of c-Jun (14), were used as negative controls (data not shown). In summary, of the eight human prostate cell lines that we examined, six revealed a down-regulation of MKK4 or c-Jun expression, and only TSU-Pr1 and LNCaP expressed all of the major components of the JNK/SAPK signaling cascade. Additional support for a connection between the SAPK/JNK cascade and metastasis suppression comes from a recent paper by Mashimo et al., which demonstrated that the expression of KAI1, the prototypical prostate cancer metastasis suppressor gene, is controlled at the transcriptional level, in part, by c-Jun (20). Therefore, taken together, our findings are consistent with the hypothesis that the interruption of the JNK/SAPK signaling cascade is an important event in the development of prostate cancer metastases.

Immunohistochemical Analysis of MKK4 Expression in Human Prostate Cancer.

To test the potential role of MKK4 down-regulation in the progression of clinical human prostate disease, we examined MKK4 expression by immunohistochemistry in normal and cancerous prostate tissues obtained during radical prostatectomy. If decreased expression of MKK4 plays a role in metastasis, we predicted MKK4 expression will correlate with the Gleason pattern, a known marker of metastatic potential.

An immunohistochemical detection method was developed to assess MKK4 specific expression in normal and cancerous tissues (Table 1). Evaluation of 16 normal prostate samples revealed strong specific staining for MKK4 in 100% of the epithelial cells in all of the samples and no staining when whole rabbit IgG was used as a negative control (Fig. 2, A and B). Therefore, the normal glands contained within each slide of the prostate cancer samples were used as an internal positive control in assessing the tumor staining (Fig. 2, C and D). An adenocarcinoma with staining intensity equivalent to the normal glandular epithelia on the same slide was assigned a staining grade of “3” and complete absence of staining was assigned a staining grade of “0” (examples seen in Fig. 2, A and D, tumor). The grade of the immunohistochemical staining was compared with the Gleason pattern of the tumor, which was assigned by a pathologist (X.Y.) with expertise in urological pathology. The Gleason grading system is the gold standard for histological analysis of prostatic adenocarcinoma in which, based on architectural patterns, a prostate tumor is assigned values between 1 and 5, with higher numbers representing more poorly differentiated and more aggressive tumors. In a clinical setting, because most prostate tumors are heterogeneous, each patient sample is assigned two separate Gleason patterns, corresponding to the two predominant histological patterns. The sum of these two Gleason patterns represents a Gleason score, ranging from 2–10. In clinical outcome studies, the Gleason score strongly correlates with the tumor stage, malignant progression, and clinical prognosis of men with prostate cancer (21). In our study, because we were interested in MKK4 expression at the cellular level, we focused on the Gleason pattern rather than the Gleason score.

MKK4 expression decreased in the more poorly differentiated tumors (Table 2; Fig. 2, C and D), supporting our hypothesis that diminished MKK4 expression would correlate with increased aggressiveness (e.g., Gleason pattern). A statistically significant, direct, inverse relationship between Gleason pattern and MKK4 expression was demonstrated. The mean staining intensities were: 1.76 (± 0.09), 1.35 (±0.10), and 0.73 (±0.16) for Gleason pattern 3, 4, and 5 lesions, respectively (Table 2). Our data demonstrated that, as the tumor pattern increased, mean MKK4 staining levels significantly decreased (P < 0.01). Furthermore, MKK4 expression remained the same or decreased as Gleason pattern increased in 36 of the 37 specimens that varied in grade.

In a recent report analyzing the expression of MKK4 in gastric cancer, patients with lower MKK4 expression had significantly better disease-free survival and overall survival (22). Although we did not examine patient outcomes directly, our findings imply that the loss of MKK4 expression will correlate with a lower disease-free survival. This hypothesis and the potential organ-specific differences in MKK4 expression in both normal and malignant tissues are the subject of ongoing investigations. Also, unlike the normal prostate, normal gastric tissue stained very weakly for MKK4. These findings imply that the expression and function of MKK4 in normal and neoplastic tissues may be context dependent. In light of these observations, we are currently examining additional cancers, including bladder, renal cell, and ovarian cancers for the relationship between MKK4 expression and metastatic propensity.

LOH Analysis of MKK4 Locus in Human Prostate Cancer Metastases.

The decrease in MKK4 expression in human prostate cancer may be due to allelic loss or mutations of the MKK4 gene. To assess this possibility, we conducted an LOH study using normal and metastatic prostate tumor samples from 24 patients; however, these samples were not amenable to immunohistochemical staining. Four polymorphic markers within the MKK4 locus were analyzed for LOH in normal and metastatic prostate tumor samples. Of the 16 specimens that were informative at the D17S969 locus, 5 samples (31%) had lost this marker, which is located within the MKK4 gene, 15,060 kbp from the p terminus. Markers distal to D17S969 (D17S954 14,007 kbp and D17S1303 14,074 kbp) and proximal to D17S969 (D17S947 16,186 kbp) were also analyzed. For D17S954, LOH was detected in 4 (40%) of 10 informative cases; for D17S1303 1 (33%) of 3 had LOH; for D17S947 2 (18%) of 11 had LOH. To exclude the possibility that the observed LOH reflected losses targeted to the nearby p53 gene, we also analyzed two markers within the p53 locus (D17S796 7,383 kbp and D17S938 7,584 kbp). In one case, LOH was restricted to the MKK4 region and did not involve the distal p53 gene.

Sequencing MKK4 exons B–K of the sample with LOH restricted to the D17S969 locus, as well as an additional case with more extensive LOH on chromosome 17, did not reveal any coding region mutations. These results suggest that the decreased expression of MKK4 in prostate cancer is not attributable to allelic loss or mutations of the MKK4 coding region. This finding is in agreement with previous studies of MKK4 and the known biology of other metastasis-suppressor genes, which are infrequently mutated (12, 13, 23).

The data from the current study support a role for the inactivation of MKK4 and its signaling cascade in the development of human prostate cancer metastases. In sum, we demonstrated that MKK4 or c-Jun, one of its signaling partners, was down-regulated in six of eight human prostate cancer cell lines, which were all derived from metastatic lesions. Because of the paucity of spontaneously immortalized human prostate primary tumor cell lines, this initial study focused on cell lines derived from metastatic lesions.

Our finding that MKK4 expression in primary prostate cancer correlates inversely with Gleason pattern supports the clinical relevance of this observation. Ongoing studies are designed to define the mechanism of MKK4 inactivation in clinical specimens, including testing the kinase activity of MKK4, which may be a more relevant measure, and testing the prognostic and clinical implications of abnormalities in the stress-signaling cascade in human prostate cancer.

Fig. 1.

MKK4 or c-Jun protein expression was absent or low in six of eight human prostate cancer cell lines derived from metastases. A, summary of major components of the kinase cascade involving MKK4. Activation of a MAPK kinase kinase (MAPKKK), such as MEKK1, leads to the subsequent activation of a MAPK kinase (MAPKK), which includes: MKK3, MKK4, MKK6, and MKK7. These activated kinases act on the MAPK proteins JNK and p38, which then modify the activity of transcription factors such as c-Jun. Analyses shown in B and C show that MKK4 or its downstream mediator c-Jun were altered in the majority of cell lines examined (shaded). B, MKK4 expression in human prostate cancer cell lines. The expression of MKK4 in control cell lines and a series of human prostate cancer cell lines was detected by immunoblotting. Rat brain, which expresses high endogenous levels of MKK4, served as a positive control, and ASPC-1 cells, which do not express MKK4 because of a homozygous deletion within the gene, served as a negative control. Primary prostatic epithelial cells grown in short-term culture were analyzed to assess protein expression as an additional control. MKK4 protein was detected using a polyclonal antibody. Actin expression was assessed as a loading control (Calbiochem). C, expression of JNK/SAPK signaling molecules in human prostate cancer cell lines. Expression of components of the MKK4 signaling cascade was examined in control cell lines and a series of human prostate cancer cell lines using immunoblotting (polyclonal antibodies). Actin expression was assessed as a loading control (Calbiochem). Representative data show expression of p38, JNK1, and c-Jun.

Fig. 1.

MKK4 or c-Jun protein expression was absent or low in six of eight human prostate cancer cell lines derived from metastases. A, summary of major components of the kinase cascade involving MKK4. Activation of a MAPK kinase kinase (MAPKKK), such as MEKK1, leads to the subsequent activation of a MAPK kinase (MAPKK), which includes: MKK3, MKK4, MKK6, and MKK7. These activated kinases act on the MAPK proteins JNK and p38, which then modify the activity of transcription factors such as c-Jun. Analyses shown in B and C show that MKK4 or its downstream mediator c-Jun were altered in the majority of cell lines examined (shaded). B, MKK4 expression in human prostate cancer cell lines. The expression of MKK4 in control cell lines and a series of human prostate cancer cell lines was detected by immunoblotting. Rat brain, which expresses high endogenous levels of MKK4, served as a positive control, and ASPC-1 cells, which do not express MKK4 because of a homozygous deletion within the gene, served as a negative control. Primary prostatic epithelial cells grown in short-term culture were analyzed to assess protein expression as an additional control. MKK4 protein was detected using a polyclonal antibody. Actin expression was assessed as a loading control (Calbiochem). C, expression of JNK/SAPK signaling molecules in human prostate cancer cell lines. Expression of components of the MKK4 signaling cascade was examined in control cell lines and a series of human prostate cancer cell lines using immunoblotting (polyclonal antibodies). Actin expression was assessed as a loading control (Calbiochem). Representative data show expression of p38, JNK1, and c-Jun.

Close modal
Fig. 2.

The MKK4 protein was expressed in the glandular epithelium of normal human prostatic tissue. A, whole rabbit IgG negative control for MKK4 staining; E, epithelial cells; S, stromal cells. B, detection of MKK4 in the glandular epithelium of normal human prostate tissue. C, representative primary human prostate Gleason pattern 4 tumor sample and normal epithelium (arrow) in the same field that was stained immunohistochemically for MKK4. D, a Gleason pattern 5 prostate tumor with an area of normal glandular epithelium (arrow) was stained immunohistochemically for MKK4.

Fig. 2.

The MKK4 protein was expressed in the glandular epithelium of normal human prostatic tissue. A, whole rabbit IgG negative control for MKK4 staining; E, epithelial cells; S, stromal cells. B, detection of MKK4 in the glandular epithelium of normal human prostate tissue. C, representative primary human prostate Gleason pattern 4 tumor sample and normal epithelium (arrow) in the same field that was stained immunohistochemically for MKK4. D, a Gleason pattern 5 prostate tumor with an area of normal glandular epithelium (arrow) was stained immunohistochemically for MKK4.

Close modal

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.

1

Supported by American Cancer Society Institutional Grant, IGR41-35-3, NIH P20 CA 66132, University of Chicago Surgery Research Committee Grant; Cancer Research Foundation Young Investigator Award and NIH First Award R29 CA69487 02 (to C. W. R-S.); National Cancer Institute (NCI)/NIH Predoctoral Cancer Biology Training Grant (to D. J. V. G.); American Foundation for Urologic Disease (to B. A. Y., C. W. R-S.); The University of Chicago RESCUE Fund (to B. A. Y., Z. D., M. A. C.); Grant-in-Aid for Scientific Research 11207029 from the Japan Society for the Promotion of Science (to Y. I., T. I.); American Cancer Society RPG-99-171-01-CCG, Department of Defense, PC991167, NIH CA73740 (to A. L.); University of Chicago Cancer Research Center-Cancer Center Support Grant, NCI 5P30 CA14599-27 (to T. K.); and Specialized Program for Research Excellence in Prostate Cancer (SPORE), NCI Grant CA69568 (to M. A. R.).

4

The abbreviations used are: MAP, mitogen-activated protein; MAPK, MAP kinase; MKK4, MAPK kinase 4; LOH, loss of heterozygosity; BRFF, Biological Research Faculty and Facility; PVP, polyvinyl pyrrolidine; ERK, extracellular signal-regulated protein kinase; JNK, c-Jun NH2-terminal protein kinase; SAPK, stress-activated protein kinase; MEK, MAP/ERK kinase; MEKK, MEK kinase.

Table 1

Immunohistochemistry controls

SamplesStaining pattern
Control cell lines  
 ASPC-1 (negative control) No staining 
 MDA PCa 2b (positive control) Nuclear staining 
Control tissues  
 Human testicle (MKK4 positive by Western) Epithelial-specific staining 
 Rat testicle (MKK4 positive by Western) Epithelial-specific staining 
 Normal human prostate (n = 16) Epithelial-specific staining 
Immunohistochemical controls  
 Isotype control No staining 
 Epitope control No staining 
 Secondary antibody only No staining 
 Endogenous peroxidase/avidin biotin No staining 
SamplesStaining pattern
Control cell lines  
 ASPC-1 (negative control) No staining 
 MDA PCa 2b (positive control) Nuclear staining 
Control tissues  
 Human testicle (MKK4 positive by Western) Epithelial-specific staining 
 Rat testicle (MKK4 positive by Western) Epithelial-specific staining 
 Normal human prostate (n = 16) Epithelial-specific staining 
Immunohistochemical controls  
 Isotype control No staining 
 Epitope control No staining 
 Secondary antibody only No staining 
 Endogenous peroxidase/avidin biotin No staining 
Table 2

Summary of MKK4 staining

To quantitate the staining of the tumor, a numeric scale of 0, 1, 2, and 3 was used, with 3 representing the staining intensity equivalent to the normal histological regions on the same slide and 0 representing no staining. In evaluating cancer samples, we used adjacent normal epithelial regions within the slide, which stained for MKK4 expression, as internal controls for evaluating the staining in the tumor region.
Gleason pattern n Mean staining SE 95% CIa 
52 1.76 0.09 (1.57–1.94) 
46 1.35 0.10 (1.16–1.55) 
14 0.73 0.16 (0.42–1.04) 
To quantitate the staining of the tumor, a numeric scale of 0, 1, 2, and 3 was used, with 3 representing the staining intensity equivalent to the normal histological regions on the same slide and 0 representing no staining. In evaluating cancer samples, we used adjacent normal epithelial regions within the slide, which stained for MKK4 expression, as internal controls for evaluating the staining in the tumor region.
Gleason pattern n Mean staining SE 95% CIa 
52 1.76 0.09 (1.57–1.94) 
46 1.35 0.10 (1.16–1.55) 
14 0.73 0.16 (0.42–1.04) 
a

CI, confidence interval.

We thank Drs. Gary Steinberg, Charles Brendler, and all of the members of our RESCUE fund effort for their unwavering, enthusiastic support of this research. We also thank Dr. Thomas Krausz and personnel in the Department of Pathology for their efforts on behalf of this research. Finally, we thank Dr. Raphael Espinosa for expertise and assistance with figure preparation and Ann Koons for her assistance in developing immunohistochemistry methods.

1
Walsh P. C., Partin A. W., Epstein J. I. Cancer control and quality of life following anatomical radical retropubic prostatectomy: results at 10 years.
J. Urol.
,
152
:
1831
-1836,  
1994
.
2
Voges G. E., McNeal J. E., Redwine E. A., Freiha F. S., Stamey T. A. Morphologic analysis of surgical margins with positive findings in prostatectomy for adenocarcinoma of the prostate.
Cancer (Phila.)
,
69
:
520
-526,  
1992
.
3
Lu-Yao G. L., McLerran D., Wasson J., Wennberg J. E. An assessment of radical prostatectomy. Time trends, geographic variation, and outcomes. The Prostate Patient Outcomes Research Team.
J. Am. Med. Assoc.
,
269
:
2633
-2636,  
1993
.
4
Christiano A. P., Yoshida B. A., Dubauskas Z., Sokoloff M., Rinker-Schaeffer C. W. Development of markers of prostate cancer metastasis: review and perspective.
Urol. Oncol.
,
5
:
217
-223,  
2000
.
5
Graefen M., Noldus J., Pichlmeier U., Haese A., Hammerer P., Fernandez S., Conrad S., Henke R., Huland E., Huland H. Early prostate-specific antigen relapse after radical retropubic prostatectomy: prediction on the basis of preoperative and postoperative tumor characteristics.
Eur. Urol.
,
36
:
21
-30,  
1999
.
6
Moul J. W. Rising PSA after local therapy failure: immediate vs. deferred treatment.
Oncology (Huntingt)
,
13
:
985
-990, 993; discussion 993995, 999,  
1999
.
7
Dillioglugil O., Leibman B. D., Kattan M. W., Seale-Hawkins C., Wheeler T. M., Scardino P. T. Hazard rates for progression after radical prostatectomy for clinically localized prostate cancer.
Urology
,
50
:
93
-99,  
1997
.
8
Partin A. W., Pound C. R., Clemens J. Q., Epstein J. I., Walsh P. C. Serum PSA after anatomic radical prostatectomy. The Johns Hopkins experience after 10 years.
Urol. Clin. North Am.
,
20
:
713
-725,  
1993
.
9
Small E. J., Reese D. M., Vogelzang N. J. Hormone-refractory prostate cancer: an evolving standard of care.
Semin. Oncol.
,
26
:
61
-67,  
1999
.
10
Hugosson J., Aus G. Natural course of localized prostate cancer. a personal view with a review of published papers.
Anticancer Res.
,
17
:
1441
-1448,  
1997
.
11
Yoshida B. A., Sokoloff M. A., Welch D. R., Rinker-Schaeffer C. W. Metastasis-suppressor genes. a review and perspective on an emerging field.
J. Natl. Cancer Inst. (Bethesda)
,
92
:
1717
-1730,  
2000
.
12
Yoshida B. A., Dubauskas Z., Chekmareva M. A., Christiano T. R., Stadler W. M., Rinker-Schaeffer C. W. Mitogen-activated protein kinase kinase 4/stress-activated protein/Erk kinase 1 (MKK4/SEK1), a prostate cancer metastasis suppressor gene encoded by human chromosome 17.
Cancer Res.
,
59
:
5483
-5487,  
1999
.
13
Teng D. H., Perry W. L., III, Hogan J. K., Baumgard M., Bell R., Berry S., Davis T., Frank D., Frye C., Hattier T., Hu R., Jammulapati S., Janecki T., Leavitt A., Mitchell J. T., Pero R., Sexton D., Schroeder M., Su P. H., Swedlund B., Kyriakis J. M., Avruch J., Bartel P., Wong A. K., Tavtigian S. V., et al Human mitogen-activated protein kinase kinase 4 as a candidate tumor suppressor.
Cancer Res.
,
57
:
4177
-4182,  
1997
.
14
Chiu R., Boyle W. J., Meek J., Smeal T., Hunter T., Karin M. The c-Fos protein interacts with c-Jun/AP-1 to stimulate transcription of AP-1 responsive genes.
Cell
,
54
:
541
-552,  
1988
.
15
Rubin M. A., Putzi M., Mucci N., Smith D. C., Wojno K., Korenchuk S., Pienta K. J. Rapid (“warm”) autopsy study for procurement of metastatic prostate cancer.
Clin. Cancer Res.
,
6
:
1038
-1045,  
2000
.
16
Zeger S. L., Liang K. Y. Longitudinal data analysis for discrete and continuous outcomes.
Biometrics
,
42
:
121
-130,  
1986
.
17
Davis R. J. Signal transduction by the c-Jun N-terminal kinase.
Biochem. Soc. Symp.
,
64
:
1
-12,  
1999
.
18
Minden A., Karin M. Regulation and function of the JNK subgroup of MAP kinases.
Biochim. Biophys. Acta
,
1333
:
F85
-F104,  
1997
.
19
Schlesinger T. K., Fanger G. R., Yujiri T., Johnson G. L. The TAO of MEKK.
Front. Biosci.
,
3
:
D1181
-1186,  
1998
.
20
Mashimo T., Bandyopadhyay S., Goodarzi G., Watabe M., Pai S. K., Gross S. C., Watabe K. Activation of the tumor metastasis suppressor gene, KAI1, by etoposide is mediated by p53 and c-Jun genes.
Biochem. Biophys. Res. Commun.
,
274
:
370
-376,  
2000
.
21
Epstein J. I., Pizov G., Walsh P. C. Correlation of pathologic findings with progression after radical retrobupic prostatectomy.
Cancer
,
71
:
3582
-3593,  
1993
.
22
Wu C. W., Li A. F., Chi C. W., Huang C. L., Shen K. H., Liu W. Y., Lin W. Human gastric cancer kinase profile and prognostic significance of MKK4 kinase.
Am. J. Pathol.
,
156
:
2007
-2015,  
2000
.
23
Su G. H., Hilgers W., Shekher M. C., Tang D. J., Yeo C. J., Hruban R. H., Kern S. E. Alterations in pancreatic, biliary, and breast carcinomas support MKK4 as a genetically targeted tumor suppressor gene.
Cancer Res.
,
58
:
2339
-2342,  
1998
.