| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Research Institute for Internal Medicine (S.L.F.), University of Oslo, N-0027 Oslo, Norway; Hormone Laboratory, Aker University Hospital (F.L., T.L., J.P.B.), University of Oslo, N-0027 Oslo, Norway; and Section of Endocrinology (S.L.F., J.B.), and Departments of Radiology (J.K.H.) and Neurosurgery (J.R.-P.), Rikshospitalet Medical Centre, N-0027 Oslo, Norway
Address all correspondence and requests for reprints to: Stine Lyngvi Fougner, Research Institute for Internal Medicine, Section of Endocrinology, Rikshospitalet Medical Centre, N-0027 Oslo, Norway. E-mail: s.l.fougner{at}medisin.uio.no.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to study RKIP levels in pituitary somatotroph adenomas, and relate them to clinical characteristics and response to octreotide treatment in patients with acromegaly.
Patients and Methods: RKIP level was analyzed by Western blot of proteins extracted from somatotroph tumors frozen a short time after surgery in 51 patients with active acromegaly. An acute somatostatin test was performed in 46 of the patients, and in 21 the IGF-I level before and 6 months after SMS treatment was available.
Results: The adenoma RKIP level correlated significantly to both the acute and the long-term octreotide responses on serum levels of GH and IGF-I, respectively. In multiple regression analyses, the RKIP level was a significant determinant for both the GH reduction in the acute test and the IGF-I reduction after approximately 6 months.
Conclusion: The RKIP level in somatotroph adenomas seems to be important for the clinical effect of SMS treatment, in which low levels of RKIP correlate to poor clinical response to SMSs.
| Introduction |
|---|
|
|
|---|
Somatostatin analogs (SMSs) are commonly used in the medical treatment of acromegaly. Two long-acting analogs are available for clinical use, octreotide and lanreotide. Both bind with high affinity to the somatostatin receptor (SSTR) subtype 2 (SSTR2), less affinity to subtype 5, and with low affinity to receptor subtype 3. Reduced tumor expression of SSTR2 is suggested to explain the lack of response to SMSs in acromegaly. Studies correlating SSTR2 mRNA levels to octreotide efficacy have had contradictory results, but most have shown a correlation between receptor mRNA level and octreotide response (3, 4, 5, 6, 7, 8). Recently, we have demonstrated a better octreotide response in patients with adenomas containing a large proportion of immunohistochemically SSTR2a-positive cells (9). However, some patients exhibited poor response despite a high proportion of SSTR2a-positive adenoma cells and vice versa.
Raf kinase inhibitory protein (RKIP) is an evolutionarily conserved and widely expressed protein. It inhibits the MAPK signaling pathway Ras/Raf-1/MEK/ERK by inhibition of Raf-1 phosphorylation and activation (10, 11, 12). This intracellular signaling pathway is involved in the control of cell proliferation and differentiation, cell death, and apoptosis (12, 13). Protein kinase C phosphorylates RKIP, which then dissociates from Raf-1, and binds to and inhibits G protein-coupled receptor kinase 2 (GRK2). This inhibition leads to decreased internalization and degradation of G protein-coupled receptors, and, therefore, a prolonged receptor signal, in addition to the enhanced signal via the MAPK pathway (14). Internalization of the G protein-coupled receptor SSTR2a is shown to be dependent of GRK2-mediated phosphorylation of the receptor (15). Moreover, RKIP inhibits TNF
-induced activation of the nuclear factor-
B pathway (16). Low levels of RKIP have been associated with an increased propensity of tumors to metastasize and were the strongest predictor for reduced survival in colorectal cancer (17, 18, 19). Loss of RKIP has also been linked to resistance to chemotherapeutic drugs in prostate and breast cancer cell lines (20).
The purpose of this study was to examine the RKIP levels in somatotroph pituitary adenomas, and relate this to tumor size and invasiveness, hormone levels, and to the clinical response to octreotide treatment.
| Subjects and Methods |
|---|
|
|
|---|
A total of 51 patients with active acromegaly were included in the study (27 females, 24 males). All underwent transsphenoidal adenomectomy at Rikshospitalet Medical Centre in the period 1996–2006. The diagnosis was based on clinical symptoms and was biochemically verified by elevated serum IGF-I compared with the normal reference range of their age and failure to suppress GH less than 2.5 mU/liter after a 75-g oral glucose load. In all patients, a pituitary tumor was visualized on a magnetic resonance imaging (MRI) scan.
One patient had previously undergone transsphenoidal surgery. There were 15 patients who received treatment with octreotide before surgery. Of these, 10 patients were randomized to 6-month preoperative octreotide treatment in ongoing clinical trials. One patient was treated preoperatively with pegvisomant. None had received radiation therapy before surgery. Table 1
provides an overview of the study population.
|
Biochemical measurements
Blood samples were drawn after an overnight fast and serum isolated. In 26 patients additional serum was stored at –80 C until analyzed. Serum IGF-I was measured by RIA (Nichols Institute, Nijmegen, The Netherlands) except in the four last included patients (Immulite; Diagnostic Products Corp., Los Angeles, CA). The same method was used when IGF-I levels were compared before and after SMS treatment. Over the years different GH immunoassays (detection limit 0.3 mU/liter by all methods) were used. At every change of method, cross-calibration was performed and, if necessary, factorial adjustments introduced. In most patients, an additional three to seven samples for measurement of GH were drawn during daytime (0800–1500 h) for calculation of mean GH level. The acute somatostatin test was performed in 46 of the patients before medical treatment with a SMS. During the test, one to three measurements of basal GH levels were performed, and two to three analyses between 2 and 4 h after sc injection of octreotide. The percent reduction of GH was calculated and nadir GH recorded for each patient.
In addition, serum IGF-I was measured in the stored samples in one run to minimize between-assay variability and was compared with previous values (r = 0.89; n = 26; P < 0.001). When available this value was used for statistical analysis.
MRI measurements
The MRI scan was available for renewed analysis in 46 patients, and in 16 patients the scans both before and after treatment with a SMS could be reevaluated. The measurements were performed by two investigators (J.K.H. and S.L.F.) blinded for clinical data. Of the T1-weighted images, the coronal view before and after contrast injection was used for the measurement of maximal transverse diameter, and the sagittal plane was used for the measurement of maximal vertical and anteroposterior diameter. For each tumor, the suprasellar, infrasellar, parasellar, anterior, and posterior (SIPAP) grading score was determined (21). The formula width x height x length x 0.5 was used to estimate tumor volume (22). As a measure for tumor extensiveness, the following grading system was used: noninvasive, maximum SIPAP 1 in the suprasellar extension; borderline invasive, maximum SIPAP 2 in the suprasellar extension or maximum 1 in the parasellar extension; and invasive, all adenomas that did not classify under the specifications for noninvasive or borderline invasive.
Protein extraction
Shortly after pituitary surgery, adenoma tissue was frozen at –70 C. The tissue was homogenized in TRIZOL Reagent (Invitrogen Corp., Carlsbad, CA), and protein was extracted following the manufacturers instructions, then precipitated and washed as described in detail previously (9). The protein concentration was measured by the Bradford dye-binding procedure using Ultrospec 3300 pro spectrophotometer (Amersham Biosciences, Buckinghamshire, UK).
Western blot
After protein wash, 8.7 µg total protein was separated on a NuPAGE 4–12% BisTris gel (Invitrogen) and blotted onto a polyvinylidene fluoride membrane. Nonfat dry milk in Tris-buffered saline with Tween 20 was used for blocking (5%) and antibody incubation (2.5%). The membranes were cut at 30–35 kDa. The part containing the smallest proteins was incubated with rabbit polyclonal anti-RKIP antibody (1:2,000; Abcam plc, Cambridge, UK) at 4 C overnight, washed three times, and incubated with the secondary antibody antigoat IgG (1:10,000; Calbiochem, San Diego, CA) for 1 h at room temperature. The same protocol was used for the membrane part containing the largest proteins but with a mouse monoclonal anti-β-actin antibody (1:10,000; Sigma-Aldrich, St. Louis, MO) and as secondary antibody an antimouse IgG (1:10,000; Jackson ImmunoResearch Laboratories, Inc., West Grove PA). This part of the membrane was then stripped by washing with 0.1 M citric acid and Tris-buffered saline (3x) with 0.05% NP-40 twice. After blocking, it was incubated with a mouse anti-
-tubulin antibody (1:1000; Sigma-Aldrich) and with the antimouse IgG as secondary antibody. The enhanced chemiluminescence Western Blotting Detection Reagents (Amersham Biosciences) were used for visualization. Protein levels were determined by the Multi Gauge software (Fujifilm Corp., Tokyo, Japan). The band signal from each antibody was adjusted for background signal, and the RKIP/β-actin and RKIP/
-tubulin ratio was calculated. For each adenoma the mean of RKIP/β-actin and RKIP/
-tubulin was designated the RKIP ratio and used in the analyses.
Most adenomas have previously been examined for SSTR2a protein level by Western blot using a polyclonal anti-SSTR2a antibody (Gramsch Laboratories, Schwabhausen, Germany), standardized against the β-actin signal. This procedure has been described in detail previously (9).
Immunohistochemistry
Semiquantitative scores of SSTR2a immunohistochemical expression in formalin-fixed, paraffin-embedded adenoma tissue sections from 47 of the patients have been presented in another study and were used in multiple regression analysis with RKIP in this study. The SSTR2a scores were based on less than 75% (grade 1 or 2) or more than 75% (grade 3) of the cells positively stained by immunohistochemistry (9).
Statistical analyses
Differences between groups were analyzed using the Mann-Whitney rank sum test and Kruskal-Wallis test when more than two groups were compared. Relationships between variables were tested by Spearman correlation analysis. Multiple linear regression analysis was performed with stepwise addition of the variables that had P values less than 0.1 in the univariate analyses. P < 0.05 was considered significant for all tests. The statistical analyses were performed using SPSS software version 13 (SPSS, Inc., Chicago, IL).
| Results |
|---|
|
|
|---|
Staining with anti-RKIP antibody gave a single band at approximately 22 kDa. As described in the Subjects and Methods section, the mean of RKIP/β-actin and RKIP/
-tubulin was used in the analyses. A representative membrane is presented in Fig. 1
.
|
The RKIP level in the somatotroph adenomas correlated to both the percent reduction of serum GH concentration during the acute octreotide test (R = 0.342; P = 0.020) and the nadir GH level during the test (R = –0.457; P = 0.001) (Fig. 2
). In a multiple regression model, both SSTR2a immunohistochemical expression scores of the adenoma and RKIP levels were significant determinants of the acute octreotide response, defined as percent GH reduction in an acute octreotide test (R2 0.401) (Table 2
).
|
|
RKIP level and the long-term octreotide response
The RKIP level in the tumor correlated to both the percent reduction in serum IGF-I concentration (R = 0.674; P = 0.001) and the absolute IGF-I reduction (R = 0.619; P = 0.003) after approximately 6-month octreotide treatment in 21 patients (Fig. 3
). Both the RKIP level and the adenoma SSTR2a immunohistochemical grading were significant determinants for the long-term percent serum IGF-I reduction in the multiple regression model (R2 0.722) (Table 2
). The good responders, defined as patients with more than 50% IGF-I reduction, had a significantly higher RKIP level (median 1.85) than the bad responders (median 0.73) (P = 0.015).
|
RKIP level and baseline patient characteristics
The RKIP level did not correlate to patient age or baseline serum GH or IGF-I levels (P = 0.41, P = 0.70). No correlation to preoperative tumor size or invasiveness could be demonstrated (P = 0.45, P = 0.84).
There was no difference in RKIP levels among the patients preoperatively treated with octreotide compared with those not pretreated (P = 0.98).
A Gs
mutation analysis was performed in 47 of the patients as described previously (9), and 20 adenomas were gsp oncogene positive. There was no difference in RKIP level between the gsp- positive and gsp-negative adenomas (P = 0.344).
RKIP level and adenoma SSTR2a status
There was no difference in RKIP level between tumors with different immunohistochemical scores for SSTR2a (P = 0.60), and no correlation between the RKIP level and the SSTR2a level measured by Western blot in a previous study (P = 0.17) (9).
| Discussion |
|---|
|
|
|---|
The only factor known to influence the efficacy of SMS treatment in acromegaly has been the tumor expression of SSTRs, mainly the subtype 2 receptor. Generally, a low receptor level correlates to a poor octreotide response with some exceptions (3, 4, 5, 6, 7, 8). However, most studies have been performed at the mRNA level, which may not necessarily correlate to the protein level of the receptor (23, 24). In our recent study of adenoma SSTR2a protein level, we demonstrated a correlation between octreotide response and adenoma SSTR2a expression immunohistochemically, in which adenomas with a high proportion of SSTR2a- positive cells responded significantly better to octreotide treatment. However, no correlation to protein level assessed by Western blot analysis of protein extracts from the tumors was found (9). Nevertheless, exceptions were observed, suggesting that other mechanisms downstream of the SSTRs may also be involved in determining the clinical effect of SMSs in the somatotroph adenomas. In the present study, we found a strong correlation between RKIP level in the adenoma assessed by Western blot and both acute and long-term clinical response to octreotide. However, the RKIP level did not correlate to the SSTR2a protein level in tumor extracts or expressed immunohistochemically. In the multiple regression models, both SSTR2a immunohistochemical grading and the RKIP level were significant determinants of the octreotide response. Patients with poor octreotide response despite a high proportion of SSTR2a-positive cells seemed to have low RKIP levels, and patients with few SSTR2a- positive cells, but relatively good octreotide response, had high RKIP levels. This suggests a model in which SSTR2a expression is a permissive factor for octreotide response in acromegaly but that RKIP is an independent modulator of the efficacy of SMSs. No correlation between RKIP level and tumor shrinkage was demonstrated, but MRI scan for this analysis was limited to 16 patients. Previous studies have not shown any correlation between biochemical effect and tumor size reduction during long-term treatment of acromegaly (25, 26).
RKIP has previously been found to be important for responses to medical treatment. Some prostate and breast cancer cell lines resistant to chemotherapeutic drugs have been shown to express low levels of RKIP. Increasing the levels by transfecting the cells with an RKIP expression vector re-sensitized the cells to drug-induced apoptosis. In drug-sensitive cell lines, down-regulation of RKIP led to the resistance to the chemotherapeutic drug (20). Rituximab up-regulates RKIP, which has been shown to sensitize non-Hodgkins lymphoma cell lines to chemotherapeutic drug-induced apoptosis (27). Both inhibition of the MEK-ERK pathway and inhibition of the nuclear factor-
B pathway were suggested as possible mechanisms for the drug-sensitizing effects.
Studies have indicated RKIP as a metastasis suppressor gene in cancer. Loss of RKIP was associated with metastasis development, and increased angiogenesis and vascular invasion were suggested as possible mechanisms (18). Somatotroph pituitary tumors rarely metastasize, but growth can be locally invasive with erosion of surrounding bone. The tumor often expands outside of the sella turcica at diagnosis, making radical surgery difficult. In our study adenoma RKIP level did not correlate to tumor size and invasiveness. This could be due to the general benign nature of these adenomas.
RKIP inhibits the MEK-ERK signaling pathway, which is also inhibited by activation of SSTR2, and SSTR subtypes 3 and 5 (28, 29). Therefore, a synergistic inhibitory effect on the MEK-ERK pathway is a possible mechanism for the association between adenoma RKIP levels and the clinical efficacy of octreotide. Phosphorylation of RKIP by protein kinase C activates RKIP binding to and inhibition of GRK2. Therefore, internalization and degradation of the SSTR are reduced. However, there was no correlation between total RKIP level and the level of SSTR2a in our study.
We found a significant correlation between the adenoma RKIP level and the clinical effect of octreotide, both in the acute test situation and after long-term treatment. The RKIP level was also a significant determinant of octreotide efficacy in the multiple regression models, in addition to the SSTR2a protein expression. However, potentially RKIP by itself might not be crucial but could be related to an associating factor of an unknown mechanism. Yet, there are several potential mechanisms for RKIP to be of importance for the SMS effect, e.g. the effect of RKIP on drug sensitivity of other cell types.
In conclusion, RKIP levels in somatotroph tumors correlated to octreotide response in patients with acromegaly. Low levels of RKIP were associated with a poor biochemical effect of the treatment, even in patients who expressed high levels of the main octreotide receptor SSTR2a. The study suggests that RKIP is an important modulator of the therapeutic effect of octreotide of GH producing tumors. Further studies are needed to reveal the mechanism for this and to evaluate the role of RKIP on SMS response in other tumors.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online January 29, 2008
Abbreviations: GRK2, G protein-coupled receptor kinase 2; MRI, magnetic resonance imaging; RKIP, Raf kinase inhibitory protein; SIPAP, suprasellar, infrasellar, parasellar, anterior, and posterior; SMS, somatostatin analog; SSTR2, somatostatin receptor subtype 2.
Received October 9, 2007.
Accepted January 18, 2008.
| References |
|---|
|
|
|---|
, Pit-1, Gi2
, and somatostatin receptor 2 in human somatotroph adenomas: involvement in octreotide sensitivity. J Clin Endocrinol Metab 84:2759–2765
B-inducing kinase and TAK1 and inhibits NF-
B activation. Mol Cell Biol 21:7207–7217
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |