NIH Public Access Author Manuscript Cancer. Author manuscript; available in PMC 2011 February 15.
NIH-PA Author Manuscript
Published in final edited form as: Cancer. 2010 February 15; 116(4): 843–851. doi:10.1002/cncr.24838.
Serum Amyloid A (SAA): a Novel Biomarker for Endometrial Cancer Emiliano Cocco, Ph.D.1, Stefania Bellone, Ph.D.1, Karim El-Sahwi, M.D.1, Marilisa Cargnelutti, Ph.D.1, Natalia Buza, M.D.2, Fattaneh A. Tavassoli, M.D.2, Peter E. Schwartz, M.D.1, Thomas J. Rutherford, M.D.1, Sergio Pecorelli, M.D.3, and Alessandro D. Santin, M.D. 1,* 1
Department of Obstetrics, Gynecology & Reproductive Sciences, Yale University School of Medicine, New Haven, CT 2
Department of Pathology, Yale University School of Medicine, New Haven, CT
3
Division of Gynecologic Oncology, University of Brescia, Brescia, Italy
NIH-PA Author Manuscript
Abstract Background—We investigated the expression of Serum-Amyloid-A (SAA) in endometrial endometrioid carcinoma (EEC), and evaluated its potential as a serum biomarker. Methods—SAA gene and protein expression levels were evaluated in EEC and normal endometrial tissues (NEC), by real time-PCR, immunohistochemistry (IHC) and flow cytometry. SAA concentration in 194 serum samples from 50 healthy-women, 42 women with benign diseases and 102 patients including 49 grade-1, 38 grade-2 and 15 grade-3 EEC was also studied by a sensitive bead-based-immunoassay.
NIH-PA Author Manuscript
Results—SAA gene expression levels were significantly higher in EEC when compared to NEC (mean-copy-number by RT-PCR = 182 vs 1.9; P=0.001). IHC revealed diffuse cytoplasmic SAA protein staining in poorly differentiated EEC tissues. High intracellular levels of SAA were identified in primary EEC cell lines evaluated by flow cytometry and SAA was found to be actively secreted in vitro. SAA concentrations (μg/ml) had medians of 6.0 in normal healthy females and 6.0 in patients with benign disease (P=0.92). In contrast, SAA values in the serum of EEC patients had a median of 23.7 significantly higher than those of the healthy group (P=0.001) and benign group (P=0.001). Patients harboring G3 EEC were found to have SAA concentrations significantly higher than G1/G2 patients. Conclusions—SAA is not only a liver-secreted-protein but is also an EEC-cell product. SAA is expressed and actively secreted by G3-EEC and it is present in high concentration in the serum of EEC patients. SAA may represent a novel biomarker for EEC to monitor disease recurrence and response to therapy. Keywords Endometrial carcinoma; Serum Amyloid A; Biomarkers; Tumor markers
*
To whom correspondence should be addressed: Alessandro D. Santin, M.D., Yale University School of Medicine, Department of Obstetrics, Gynecology & Reproductive Sciences Rm. 305 LSOG, 333 Cedar Street; PO Box 208063, New Haven, CT, 06520-8063. Phone: 203-737-4450. Fax; 203-737-4339.
[email protected].
Cocco et al.
Page 2
INTRODUCTION NIH-PA Author Manuscript
Cancer of the uterine corpus is the most prevalent gynecologic tumor in women, with an estimated 40,100 cases and 7470 deaths in the United States in 2008 (1). Based on both clinical and histopathological variables, two subtypes of endometrial carcinoma (EC), namely Type I and Type II tumors, have been described (2). Type I ECs, which account for the majority of cases, are estrogen-related tumors usually well differentiated and endometrioid in histology. Typically these patients have a favorable prognosis with appropriate therapy. In contrast, Type II ECs include poorly differentiated endometrioid tumors (G3-EEC), serous papillary and clear cells ECs. These tumors are not associated with hyperestrogenic factors and they are more likely to have deep myometrial invasion and/or metastases at presentation and often recur despite aggressive clinical interventions (3). G3EEC accounts for the majority of Type II ECs and unfortunately, to date, no good marker for screening or disease monitoring for these biologically aggressive cancers is available. In this regard, CA125 is often used in clinical practice to monitor EC patients (4). However, this marker appears to have limited utility in monitoring the effects of adjuvant therapy or in the prediction of tumor recurrence (5). The discovery of novel diagnostic and therapeutic markers against this aggressive subset of endometrial cancers remains a high priority.
NIH-PA Author Manuscript
Human serum Amyloid A (SAA), is an HDL-associated lipoprotein known to play a major role as a modulator of inflammation and in the metabolism and transport of cholesterol (6). Of interest, SAA has been recently proposed as a potentially useful biomarker to monitor patients harboring human tumors including gastric and nasopharyngeal cancer (7,8). Moreover, in lung cancer patients, using mass spectrometry and proteomic technologies, SAA was identified as the top differentially expressed protein able to differentiate the serum of patients from the serum of healthy individuals (9). One major problem with the use of SAA, an acute phase reactant, as a potential serum marker in human cancer patients, is the fact that its elevation in the serum of patients is suggested to be of liver origin rather than a tumor-cell product (10). Indeed, SAA level in the blood may elevate up to 1000-fold in response of the body to various injuries including trauma and various inflammations in addition to neoplasia (10). Importantly, however, extrahepatic SAA expression has been previously demonstrated in several histologically normal tissues, predominantly by their epithelium (11,12). Unfortunately, only scant information regarding SAA expression in malignant human tissues has been so far reported and, to our knowledge, no studies have yet addressed a potential direct secretion of SAA by human endometrial tumors. This report represents the first investigation examining SAA expression and secretion in human endometrial carcinoma.
NIH-PA Author Manuscript
PATIENTS AND METHODS Primary Tumors Snap frozen tumor biopsies and tumor samples were derived from primary specimens staged according to the F.I.G.O. (1988) surgical staging system. Only specimens with > 75% tumor content were used in the RT-PCR experiments. Briefly, fresh tumor biopsies from 18 EEC [obtained from 5 FIGO grade 1 (G1), 10 FIGO grade 2 (G2) and 3 FIGO grade 3 (G3) EEC patients (age 63± 9: mean ± SD)] were obtained under approval of the Institutional Review Board at the time of surgery and analyzed for SAA expression. Patients from which fresh tumor biopsies were obtained included 9 stage I, 5 stage II and 4 stage III patients. Total abdominal hysterectomy, bilateral salpingo-oophorectomy, lymph node dissection and washings were performed in all endometrial cancer patients. Normal endometrial control cell samples (NEC) were obtained from biopsies of benign hysterectomy specimens obtained from women of similar age. Three primary EEC cell lines derived from patients harboring G2/G3 endometrioid carcinomas (i.e., EEC-ARK-1, EEC-ARK-2 and EECCancer. Author manuscript; available in PMC 2011 February 15.
Cocco et al.
Page 3
NIH-PA Author Manuscript
ARK-3), were also established as short term cultures following previously reported standard tissue culture techniques (13). Briefly, tumor tissues obtained from cancer patients were mechanically minced and enzymatically dissociated with 0.14% collagenase Type I (Sigma, St. Louis, MO) and 0.01% DNAse (Sigma, 2000 KU/mg) in RPMI 1640 media, as described previously by Santin et al, (13). After 1–2 hrs incubation with enzyme on a magnetic stirring apparatus at 37°C in an atmosphere of 5% CO2, the resulting suspension was collected by centrifugation at 100 g for 5–10 minutes and washed twice with RPMI 1640 medium (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum (FBS, Gemini, Woodland, CA). The final pellet was then placed in RPMI 1640 (Invitrogen) containing 10% FBS, 200 U/ml penicillin, and 200 μg/ml streptomycin in tissue culture flasks or Petri dishes (Corning, Acton, MA). The epithelial nature and the purity of EEC cultures was verified by immunohistochemical staining and flow cytometric analysis with antibodies against cytokeratin and vimentin as previously described (13). Only primary cultures which had at least 90% viability and contained >99% epithelial cells were used for SAA quantification by a sensitive bead-based immunoassay, as described below. RNA extraction and quantitative real-time PCR
NIH-PA Author Manuscript
RNA isolation from all primary snap frozen samples including eighteen EEC as well as three normal endometrial cell controls was performed using TRIzol Reagent (Invitrogen) according to the manufacturer’s instructions. Quantitative PCR was done with a 7500 Real Time PCR System using the manufacturer’s recommended protocol (Applied Biosystems)to evaluate expression SAA in all the samples. Each reaction was run in triplicate. Briefly, 5 μg of total RNA from each sample was reverse transcribed using SuperScript III first-strand cDNA synthesis (Invitrogen). Five microliters of reverse transcribed RNA samples (from 500 μL of total volume) were amplified by using the TaqMan Universal PCR Master Mix (Applied Biosystems) to produce PCR products specific for SAA. The primers for SAA were obtained from Applied Biosystems (Assay ID Hs00761940_s1). The comparative threshold cycle (CT) method (Applied Biosystems) was used to determine gene expression in each sample relative to the value observed in the lowest nonmalignant endometrial epithelial cell sample, using glyceraldehyde-3-phosphatedehydrogenase (Assay ID Hs99999905_m1) RNA as internal controls.
NIH-PA Author Manuscript
Intracellular Flow cytometry—The mouse anti-human anti-SAA monoclonal antibody (i.e., clone mcl, DAKO Corporation; Carpinteria, CA), was used for our flow cytometry study. Briefly, freshly established EEC cell lines and control cells were fixed with 2% paraformaldehyde in PBS, washed and permeabilized by incubation in PBS plus 1% BSA and 0.5% saponin (S-7900, Sigma) for 10 min at room temperature. Tumor cells were stained with anti-SAA MAb and isotype-matched controls (DAKO Corporation; Carpinteria, CA). After staining, cells were washed twice with PBS plus 0.5% BSA. Secondary goatanti-mouse antibody (IgG1-FITC, cat# 349031, Beckton Dickinson) was then added for 30 min at 4 °C. Cells were then washed twice with PBS plus 0.5% BSA. Analysis was conducted with a FACScalibur utilizing CellQuest™ software (Beckton Dickinson). SAA Immunostaining of Formalin-fixed Tumor Tissues—Formalin-fixed paraffinembedded normal endometrial control tissues and tumor tissues were evaluated by standard immunohistochemical staining for SAA expression. Study blocks (i.e., 10 samples including 5 EEC and 5 NEC) were selected after histopathologic review by a surgical pathologist. The most representative block was selected for each specimen. Briefly, deparaffinized and rehydrated sections were treated according to the manufacturer’s instructions (DAKO). The antibody was diluted 1:20 in DAKO Antibody Diluent (DAKO Corporation) and incubated at pH 9.0 for 1/2 hr at room temperature. DAKO Envision™ system was used for secondary detection and color was developed using DAB chromogen (DAKO) for 5 min followed by
Cancer. Author manuscript; available in PMC 2011 February 15.
Cocco et al.
Page 4
NIH-PA Author Manuscript
counterstaining with hematoxylin. The anti-SAA monoclonal antibody used (i.e., clone mcl, DAKO Corporation; Carpinteria, CA), was directed against AA-amyloid fibril protein. The preparation and specificity of this antibody has been previously described and demonstrated (14). Negative controls included replacement of the primary antibodies by PBS and by normal mouse isotype matched IgG (IgG2a, kappa; DAKO Corporation). Liver sections were used as positive controls. Analysis of SAA secretion in tumor samples To evaluate the potential secretion of SAA by primary EEC, supernatants obtained from EEC-ARK-1, EEC-ARK-2 and EEC-ARK-3 as well as multiple control cell lines including normal human fibroblasts, EBV-transformed B cells (LCL) and cervical carcinoma cell lines were evaluated by a sensitive bead-based immunoassay (Millipore, Corp. Danvers, MA). Briefly, tumor supernatants to be tested for SAA secretion were collected by primary tumor cell lines seeded at a density of 1 × 105 cells/ml in tissue culture Petri dishes (Corning) in RPMI 1640 media, supplemented with 10% FBS (i.e., EEC and human fibroblasts), or serum-free keratinocyte medium (KFSM, i.e., cervical cancer cell lines). After 72 hrs incubation at 37°C, supernatants were aspirated, rendered cell-free by centrifugation at 1,500 rpm for 10 minutes, and stored at −20°C before being analyzed for SAA by a beadbased immunoassay (see below).
NIH-PA Author Manuscript
Measurement of SAA concentration in serum samples
NIH-PA Author Manuscript
SAA concentration was quantified in the serum of 50 apparently healthy women, 42 women with benign diseases (i.e., 22 uterine fibroids, 8 ovarian cysts and 12 endometrial polyps), and 102 women with histologically proven primary EEC, by a commercially available beadbased immunoassay (Lincoplex kit, acute phase proteins, Millipore, Corp. Danvers, MA). Patient characteristics are described in Table 1. All samples tested were derived from patients who had provided written informed consent and the study was approved by the local institutional review boards. In brief, the assay is based on conventional sandwich assay technology. The antibody specific to SAA is covalently coupled to Luminex microspheres. The microspheres are incubated with standards, controls, and samples (25 μl) in a 96wellmicrotiter filter plate for 1 h at room temperature. After incubation, the plate is washed to remove excess reagents, and detection antibody is added. After 30-min incubation at room temperature, streptavidin-phycoerythrin is added for an additional 30 min. After a final wash, the beads are resuspended in buffer and read on a Bio-Rad Luminex100 Instrument to determine the concentration of SAA. All specimens were tested in replicate wells. Results are reported as the mean of the replicates. Serum samples from all patients were collected before surgery and stored at −80°C until analysis. Detailed information about this assay is available at: http://www.millipore.com/userguides.nsf/ dda0cb48c91c0fb6852567430063b5d6/018e9e0e26525c1185257259004e227e/$FILE/ hcvd2-67bk.pdf. Statistical Analysis For q-RT-PCR data, the right-skewing was removed by taking copy-number ratios relative to the lowest-expressing NEC sample (“relative copy numbers”), log2-transforming them to ΔCTs, and comparing the results via unequal-variance t-test for the EEC-versus-NEC difference. The analyses of differences among supernatants obtained from tumor cultures with different histologies, and among expression levels measured by flow cytometry and IHC, were performed using the Wilcoxon-Mann-Whitney (WMW) test. SAA serum concentrations among the different groups of patients (i.e., healthy controls, benign gynecologic diseases, and EEC with different degree of differentiation) were summarized as medians and ranges, and compared for pairwise differences via the WMW test. SPSS 15 Cancer. Author manuscript; available in PMC 2011 February 15.
Cocco et al.
Page 5
(SPSS Inc., Chicago, IL) was used for statistical analysis. A 5% significance level was used for all statistical comparisons.
NIH-PA Author Manuscript
RESULTS SAA Expression in Snap Frozen EEC by Quantitative Real-Time PCR To minimize the risk of contamination of EEC RNA with that of normal cells or endometrial tumor cells with different histology (i.e., clear cells or uterine serous tumors), we extracted RNA to be evaluated for SAA expression by RT-PCR from eighteen pure EEC containing a minimum of 75% tumor cells. A comparison of the q-RT-PCR data for SAA in EEC versus NEC as controls is shown in Figure 1. Significant expression differences between EEC and NEC were readily apparent (Figure 1). Relative copy numbers in NEC control samples had a mean ± SEM of 1.9 ± 0.6 and ranged from 1.0 to 3.1. By contrast, relative copies in EEC samples had a mean ± SEM of 182 ± 69 and ranged from 1.7 to 1090. The fold change in mean relative copy numbers was 95.8 (Figure 1; P=0.01).
NIH-PA Author Manuscript
Intracellular EEC expression in EEC cell lines by Flow cytometry—To determine whether the high expression of SAA gene detected by q-RT-PCR assays in flash frozen EEC also results in high expression of the SAA protein, we performed intracellular flowcytometry analysis of SAA protein expression in 3 primary EEC established as short-term cultures in vitro in our laboratory. As shown in Table 2, all 3 primary EEC-culture cell lines were found positive for intracellular SAA expression by flow cytometry (i.e., 100% positive cells; mean fluorescence intensity [MFI] range from 26 to 65) (Table 2). In contrast, significantly lower expression of SAA was detected in Epstein-Barr transformed B cells (LCL) and cervical cancer cell lines (CVX) used as controls (i.e., 65–71% positive cells; MFI range from 11 to 20 and 69–84% positive cells; MFI range from 9 to 15, respectively) by flow cytometry (P=0.03 for both EEC vs LCL and EEC vs CVX, Table 2). SAA Expression by Immunohistochemistry in EEC
NIH-PA Author Manuscript
To evaluate whether the high SAA expression detected by flow cytometry on primary EEC cell lines was comparable to the expression of SAA of the EEC specimens from which EEC primary tumor cell lines were derived and/or whether in vitro expansion conditions may have modified protein expression, we evaluated SAA by immunohistochemical staining on formalin fixed tumor tissue. As representatively shown in Figure 2 for EEC-ARK-1 as well as another unrelated G3 EEC, cytoplasmic positivity for SAA was detected by IHC in all tissues derived from poorly differentiated (G3) EEC. In contrast, no SAA positivity was detected in well differentiated (G1) and moderately differentiated (G2) EEC (Figure 2). The intensity of staining for SAA in G3 EEC was significantly higher when compared to normal endometrial controls and G1 and G2 EEC (P 99
100
65
EEC3-ARK3
Endometrioid
> 99
100
43
CVX-1
squamous
> 99
79
9
CVX-2
squamous
> 99
69
11
CVX-3
squamous
> 99
84
15
LCL-1
B cells
0
65
20
LCL-2
B cells
0
71
11
NIH-PA Author Manuscript NIH-PA Author Manuscript Cancer. Author manuscript; available in PMC 2011 February 15.
Cocco et al.
Page 15
Table 3
Serum SAA levels in non cancer (healthy), benign disease and EEC patients
NIH-PA Author Manuscript
Subject Group
N
Mean ± SEM*
Median
Healthy
50
14.9 ± 3.6
6.0
Benign disease
42
17.6 ± 5.3
6.0
EEC
102
74.1 ± 23.1
23.7
G1
49
39.7 ± 8.0
18.3
G2
38
47.9 ± 9.5
24.5
G3
15
252.6 ± 148.9
68.9
Pairwise Comparison
P value‡
Healthy versus Benign
0.917
Healthy versus EEC
0.0001
Benign versus EEC
0.0001
G1/G2 versus G3
0.02
NIH-PA Author Manuscript
*
Mean ± standard error (serum SAA is reported in μg/mL).
‡
P values are from the two-sided Wilcoxon-Mann-Whitney test of the indicated comparison.
NIH-PA Author Manuscript Cancer. Author manuscript; available in PMC 2011 February 15.
